[codex] Treat max as a first-class reasoning effort (#30467)

## Why

The Bedrock GPT-5.6 catalog advertises `max`, but Codex treated it as an
opaque custom effort. That made the reasoning picker render it as
lowercase `max` while known efforts use productized labels.

Making `max` a known effort aligns catalog data, parsing, and UI
presentation without changing the `max` wire value or persisted
representation.

## What changed

- Add first-class `ReasoningEffort::Max` parsing and serialization.
- Use the typed effort in the Bedrock catalog and render it as `Max` in
the TUI.
- Preserve forward-compatible custom-effort coverage with a genuinely
unknown `future` value.

### Before
<img width="559" height="124" alt="Screenshot 2026-06-28 at 12 08 47 PM"
src="https://github.com/user-attachments/assets/7c43cf4f-020b-4605-9239-0a9c97eb7364"
/>

### After
<img width="558" height="107" alt="Screenshot 2026-06-28 at 12 09 10 PM"
src="https://github.com/user-attachments/assets/b9cc5ded-c940-43b4-b024-bba25abe0a17"
/>

codex-rs/core/src/client.rs

@@ -169,7 +169,7 @@ pub(crate) struct CompactConversationRequestSettings {
 
 fn reasoning_effort_for_request(effort: ReasoningEffortConfig) -> ReasoningEffortConfig {
     match effort {
-        ReasoningEffortConfig::Ultra => ReasoningEffortConfig::Custom("max".to_string()),
+        ReasoningEffortConfig::Ultra => ReasoningEffortConfig::Max,
         effort => effort,
     }
 }

codex-rs/core/src/client_tests.rs

@@ -292,10 +292,7 @@ fn ultra_reasoning_uses_max_for_requests() {
             super::reasoning_effort_for_request(ReasoningEffort::Ultra),
             super::reasoning_effort_for_request(ReasoningEffort::High),
         ),
-        (
-            ReasoningEffort::Custom("max".to_string()),
-            ReasoningEffort::High,
-        )
+        (ReasoningEffort::Max, ReasoningEffort::High,)
     );
 }
 

codex-rs/core/tests/suite/client.rs

@@ -2171,7 +2171,7 @@ async fn skills_use_aliases_in_developer_message_under_budget_pressure() {
 }
 
 #[tokio::test(flavor = "multi_thread", worker_threads = 2)]
-async fn includes_configured_effort_in_request() -> anyhow::Result<()> {
+async fn includes_configured_max_effort_in_request() -> anyhow::Result<()> {
     skip_if_no_network!(Ok(()));
     let server = MockServer::start().await;
 
@@ -2183,7 +2183,7 @@ async fn includes_configured_effort_in_request() -> anyhow::Result<()> {
     let TestCodex { codex, .. } = test_codex()
         .with_model("gpt-5.4")
         .with_config(|config| {
-            config.model_reasoning_effort = Some(ReasoningEffort::Medium);
+            config.model_reasoning_effort = Some(ReasoningEffort::Max);
         })
         .build(&server)
         .await?;
@@ -2212,7 +2212,7 @@ async fn includes_configured_effort_in_request() -> anyhow::Result<()> {
             .get("reasoning")
             .and_then(|t| t.get("effort"))
             .and_then(|v| v.as_str()),
-        Some("medium")
+        Some("max")
     );
 
     Ok(())

codex-rs/core/tests/suite/remote_models.rs

@@ -329,7 +329,7 @@ async fn remote_models_long_model_slug_is_sent_with_custom_reasoning() -> Result
         /*priority*/ 1_000,
         TruncationPolicyConfig::bytes(/*limit*/ 10_000),
     );
-    let custom_reasoning_effort = ReasoningEffort::Custom("max".to_string());
+    let custom_reasoning_effort = ReasoningEffort::Custom("future".to_string());
     remote_model.default_reasoning_level = Some(custom_reasoning_effort.clone());
     remote_model.supported_reasoning_levels = vec![
         ReasoningEffortPreset {
@@ -391,7 +391,7 @@ async fn remote_models_long_model_slug_is_sent_with_custom_reasoning() -> Result
         .and_then(|reasoning| reasoning.get("summary"))
         .and_then(|value| value.as_str());
     assert_eq!(body["model"].as_str(), Some(requested_model));
-    assert_eq!(reasoning_effort, Some("max"));
+    assert_eq!(reasoning_effort, Some("future"));
     assert_eq!(reasoning_summary, Some("detailed"));
 
     Ok(())