Remove reasoning format (#8484)

This isn't very useful parameter. 

logic:
```
if model puts `**` in their reasoning, trim it and visualize the header.
if couldn't trim: don't render
if model doesn't support: don't render
```

We can simplify to:
```
if could trim, visualize header.
if not, don't render
```

codex-rs/core/src/config/mod.rs

@@ -38,7 +38,6 @@ use codex_protocol::config_types::SandboxMode;
 use codex_protocol::config_types::TrustLevel;
 use codex_protocol::config_types::Verbosity;
 use codex_protocol::openai_models::ReasoningEffort;
-use codex_protocol::openai_models::ReasoningSummaryFormat;
 use codex_rmcp_client::OAuthCredentialsStoreMode;
 use codex_utils_absolute_path::AbsolutePathBuf;
 use codex_utils_absolute_path::AbsolutePathBufGuard;
@@ -303,9 +302,6 @@ pub struct Config {
     /// Optional override to force-enable reasoning summaries for the configured model.
     pub model_supports_reasoning_summaries: Option<bool>,
 
-    /// Optional override to force reasoning summary format for the configured model.
-    pub model_reasoning_summary_format: Option<ReasoningSummaryFormat>,
-
     /// Optional verbosity control for GPT-5 models (Responses API `text.verbosity`).
     pub model_verbosity: Option<Verbosity>,
 
@@ -786,9 +782,6 @@ pub struct ConfigToml {
     /// Override to force-enable reasoning summaries for the configured model.
     pub model_supports_reasoning_summaries: Option<bool>,
 
-    /// Override to force reasoning summary format for the configured model.
-    pub model_reasoning_summary_format: Option<ReasoningSummaryFormat>,
-
     /// Base URL for requests to ChatGPT (as opposed to the OpenAI API).
     pub chatgpt_base_url: Option<String>,
 
@@ -1379,7 +1372,6 @@ impl Config {
                 .or(cfg.model_reasoning_summary)
                 .unwrap_or_default(),
             model_supports_reasoning_summaries: cfg.model_supports_reasoning_summaries,
-            model_reasoning_summary_format: cfg.model_reasoning_summary_format.clone(),
             model_verbosity: config_profile.model_verbosity.or(cfg.model_verbosity),
             chatgpt_base_url: config_profile
                 .chatgpt_base_url
@@ -3189,7 +3181,6 @@ model_verbosity = "high"
                 model_reasoning_effort: Some(ReasoningEffort::High),
                 model_reasoning_summary: ReasoningSummary::Detailed,
                 model_supports_reasoning_summaries: None,
-                model_reasoning_summary_format: None,
                 model_verbosity: None,
                 chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
                 base_instructions: None,
@@ -3273,7 +3264,6 @@ model_verbosity = "high"
             model_reasoning_effort: None,
             model_reasoning_summary: ReasoningSummary::default(),
             model_supports_reasoning_summaries: None,
-            model_reasoning_summary_format: None,
             model_verbosity: None,
             chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
             base_instructions: None,
@@ -3372,7 +3362,6 @@ model_verbosity = "high"
             model_reasoning_effort: None,
             model_reasoning_summary: ReasoningSummary::default(),
             model_supports_reasoning_summaries: None,
-            model_reasoning_summary_format: None,
             model_verbosity: None,
             chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
             base_instructions: None,
@@ -3457,7 +3446,6 @@ model_verbosity = "high"
             model_reasoning_effort: Some(ReasoningEffort::High),
             model_reasoning_summary: ReasoningSummary::Detailed,
             model_supports_reasoning_summaries: None,
-            model_reasoning_summary_format: None,
             model_verbosity: Some(Verbosity::High),
             chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
             base_instructions: None,

codex-rs/core/src/models_manager/manager.rs

@@ -354,7 +354,6 @@ mod tests {
             "truncation_policy": {"mode": "bytes", "limit": 10_000},
             "supports_parallel_tool_calls": false,
             "context_window": null,
-            "reasoning_summary_format": "none",
             "experimental_supported_tools": [],
         }))
         .expect("valid model")

codex-rs/core/src/models_manager/model_family.rs

@@ -3,7 +3,6 @@ use codex_protocol::openai_models::ApplyPatchToolType;
 use codex_protocol::openai_models::ConfigShellToolType;
 use codex_protocol::openai_models::ModelInfo;
 use codex_protocol::openai_models::ReasoningEffort;
-use codex_protocol::openai_models::ReasoningSummaryFormat;
 
 use crate::config::Config;
 use crate::truncate::TruncationPolicy;
@@ -48,9 +47,6 @@ pub struct ModelFamily {
     // The reasoning effort to use for this model family when none is explicitly chosen.
     pub default_reasoning_effort: Option<ReasoningEffort>,
 
-    // Define if we need a special handling of reasoning summary
-    pub reasoning_summary_format: ReasoningSummaryFormat,
-
     /// Whether this model supports parallel tool calls when using the
     /// Responses API.
     pub supports_parallel_tool_calls: bool,
@@ -88,9 +84,6 @@ impl ModelFamily {
         if let Some(supports_reasoning_summaries) = config.model_supports_reasoning_summaries {
             self.supports_reasoning_summaries = supports_reasoning_summaries;
         }
-        if let Some(reasoning_summary_format) = config.model_reasoning_summary_format.as_ref() {
-            self.reasoning_summary_format = reasoning_summary_format.clone();
-        }
         if let Some(context_window) = config.model_context_window {
             self.context_window = Some(context_window);
         }
@@ -128,7 +121,6 @@ impl ModelFamily {
             truncation_policy,
             supports_parallel_tool_calls,
             context_window,
-            reasoning_summary_format,
             experimental_supported_tools,
         } = model;
 
@@ -144,7 +136,6 @@ impl ModelFamily {
         self.truncation_policy = truncation_policy.into();
         self.supports_parallel_tool_calls = supports_parallel_tool_calls;
         self.context_window = context_window;
-        self.reasoning_summary_format = reasoning_summary_format;
         self.experimental_supported_tools = experimental_supported_tools;
     }
 
@@ -175,7 +166,6 @@ macro_rules! model_family {
             context_window: Some(CONTEXT_WINDOW_272K),
             auto_compact_token_limit: None,
             supports_reasoning_summaries: false,
-            reasoning_summary_format: ReasoningSummaryFormat::None,
             supports_parallel_tool_calls: false,
             apply_patch_tool_type: None,
             base_instructions: BASE_INSTRUCTIONS.to_string(),
@@ -250,7 +240,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_CODEX_INSTRUCTIONS.to_string(),
             experimental_supported_tools: vec![
                 "grep_files".to_string(),
@@ -270,7 +259,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_2_CODEX_INSTRUCTIONS.to_string(),
             apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
             shell_type: ConfigShellToolType::ShellCommand,
@@ -299,7 +287,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_2_CODEX_INSTRUCTIONS.to_string(),
             apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
             shell_type: ConfigShellToolType::ShellCommand,
@@ -312,7 +299,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_2_CODEX_INSTRUCTIONS.to_string(),
             apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
             shell_type: ConfigShellToolType::ShellCommand,
@@ -325,7 +311,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_1_CODEX_MAX_INSTRUCTIONS.to_string(),
             apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
             shell_type: ConfigShellToolType::ShellCommand,
@@ -341,7 +326,6 @@ pub(super) fn find_family_for_model(slug: &str) -> ModelFamily {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             base_instructions: GPT_5_CODEX_INSTRUCTIONS.to_string(),
             apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
             shell_type: ConfigShellToolType::ShellCommand,
@@ -416,7 +400,6 @@ fn derive_default_model_family(model: &str) -> ModelFamily {
         context_window: None,
         auto_compact_token_limit: None,
         supports_reasoning_summaries: false,
-        reasoning_summary_format: ReasoningSummaryFormat::None,
         supports_parallel_tool_calls: false,
         apply_patch_tool_type: None,
         base_instructions: BASE_INSTRUCTIONS.to_string(),
@@ -460,7 +443,6 @@ mod tests {
             truncation_policy: TruncationPolicyConfig::bytes(10_000),
             supports_parallel_tool_calls: false,
             context_window: None,
-            reasoning_summary_format: ReasoningSummaryFormat::None,
             experimental_supported_tools: Vec::new(),
         }
     }
@@ -524,7 +506,6 @@ mod tests {
             experimental_supported_tools: vec!["local".to_string()],
             truncation_policy: TruncationPolicy::Bytes(10_000),
             context_window: Some(100),
-            reasoning_summary_format: ReasoningSummaryFormat::None,
         );
 
         let updated = family.with_remote_overrides(vec![ModelInfo {
@@ -549,7 +530,6 @@ mod tests {
             truncation_policy: TruncationPolicyConfig::tokens(2_000),
             supports_parallel_tool_calls: true,
             context_window: Some(400_000),
-            reasoning_summary_format: ReasoningSummaryFormat::Experimental,
             experimental_supported_tools: vec!["alpha".to_string(), "beta".to_string()],
         }]);
 
@@ -568,10 +548,6 @@ mod tests {
         assert_eq!(updated.truncation_policy, TruncationPolicy::Tokens(2_000));
         assert!(updated.supports_parallel_tool_calls);
         assert_eq!(updated.context_window, Some(400_000));
-        assert_eq!(
-            updated.reasoning_summary_format,
-            ReasoningSummaryFormat::Experimental
-        );
         assert_eq!(
             updated.experimental_supported_tools,
             vec!["alpha".to_string(), "beta".to_string()]