2- Use string service tiers in session protocol (#20971)

## Summary
- break service tier session/op/app-server protocol fields from the
closed enum to string tier ids
- send the service tier string directly through model requests, prewarm,
compaction, memories, and TUI/app-server turn starts
- regenerate app-server protocol JSON/TypeScript schemas, removing the
standalone ServiceTier TS enum

## Verification
- just fmt
- cargo check -p codex-core -p codex-app-server -p codex-tui
- just write-app-server-schema

---------

Co-authored-by: Codex <noreply@openai.com>

codex-rs/memories/write/src/runtime.rs

@@ -18,7 +18,6 @@ use codex_otel::TelemetryAuthMode;
 use codex_protocol::SessionId;
 use codex_protocol::ThreadId;
 use codex_protocol::config_types::ReasoningSummary;
-use codex_protocol::config_types::ServiceTier;
 use codex_protocol::openai_models::ModelInfo;
 use codex_protocol::openai_models::ReasoningEffort;
 use codex_protocol::protocol::InitialHistory;
@@ -46,7 +45,7 @@ pub(crate) struct StageOneRequestContext {
     pub(crate) session_telemetry: SessionTelemetry,
     pub(crate) reasoning_effort: Option<ReasoningEffort>,
     pub(crate) reasoning_summary: ReasoningSummary,
-    pub(crate) service_tier: Option<ServiceTier>,
+    pub(crate) service_tier: Option<String>,
     pub(crate) turn_metadata_header: Option<String>,
 }
 
@@ -194,7 +193,7 @@ impl MemoryStartupContext {
                 &context.session_telemetry,
                 context.reasoning_effort,
                 context.reasoning_summary,
-                context.service_tier,
+                context.service_tier.clone(),
                 context.turn_metadata_header.as_deref(),
                 &InferenceTraceContext::disabled(),
             )

codex-rs/memories/write/src/startup_tests.rs

@@ -253,14 +253,18 @@ async fn memories_startup_phase1_uses_live_thread_service_tier() -> anyhow::Resu
             model: None,
             effort: None,
             summary: None,
-            service_tier: Some(Some(ServiceTier::Fast)),
+            service_tier: Some(Some(ServiceTier::Fast.request_value().to_string())),
             collaboration_mode: None,
             personality: None,
         })
         .await?;
 
-    let config_snapshot = wait_for_service_tier(&test, Some(ServiceTier::Fast)).await?;
-    assert_eq!(config_snapshot.service_tier, Some(ServiceTier::Fast));
+    let config_snapshot =
+        wait_for_service_tier(&test, Some(ServiceTier::Fast.request_value().to_string())).await?;
+    assert_eq!(
+        config_snapshot.service_tier,
+        Some(ServiceTier::Fast.request_value().to_string())
+    );
 
     let context = crate::runtime::MemoryStartupContext::new(
         Arc::clone(&test.thread_manager),
@@ -277,7 +281,10 @@ async fn memories_startup_phase1_uses_live_thread_service_tier() -> anyhow::Resu
             ReasoningEffort::Low,
         )
         .await;
-    assert_eq!(request_context.service_tier, Some(ServiceTier::Fast));
+    assert_eq!(
+        request_context.service_tier,
+        Some(ServiceTier::Fast.request_value().to_string())
+    );
 
     shutdown_test_codex(&test).await?;
     Ok(())
@@ -394,7 +401,7 @@ async fn wait_for_request(mock: &ResponseMock, expected_count: usize) -> Vec<Res
 
 async fn wait_for_service_tier(
     test: &TestCodex,
-    expected_service_tier: Option<ServiceTier>,
+    expected_service_tier: Option<String>,
 ) -> anyhow::Result<codex_core::ThreadConfigSnapshot> {
     let deadline = Instant::now() + Duration::from_secs(10);
     loop {