Remove retired model entries while preserving migration prompts (#44250)

## Why

Saved model selections can outlive their bundled catalog entries and still need migration guidance.

## What changed

- Remove `gpt-5.2` and `gpt-5.4-mini` from the bundled model catalog.
- Preserve migration from saved `gpt-5.4-mini` selections to `gpt-5.6-luna` when the old model is absent, while respecting catalog-provided migration metadata.
- Update tests and snapshots to use current models or explicit test fixtures, and remove the obsolete prompt fixture.

## Testing

Extend migration tests to cover catalog metadata precedence, saved selections, replacement model and reasoning-effort persistence, acknowledged prompts, and missing or hidden migration targets.

GitOrigin-RevId: 82beea504cc4a15699677c28a61ec0939af3069c

codex-rs/core/src/session/step_activation_tests.rs

@@ -57,8 +57,8 @@ fn activation_models() -> Vec<ModelInfo> {
         .expect("bundled models")
         .models
         .into_iter()
-        .find(|model| model.slug == "gpt-5.4")
-        .expect("bundled gpt-5.4");
+        .find(|model| model.slug == "gpt-5.5")
+        .expect("bundled gpt-5.5");
     [MODEL_A, MODEL_B]
         .into_iter()
         .map(|slug| ModelInfo {

codex-rs/core/src/session/tests.rs

@@ -9,7 +9,6 @@ use crate::agents_md_manager::AgentsMdManager;
 use crate::compact::InitialContextInjection;
 use crate::config::ConfigBuilder;
 use crate::config::ConfigOverrides;
-use crate::config::test_config;
 use crate::context::ContextualUserFragment;
 use crate::context::DeveloperInstructions;
 use crate::context::GuardianContextMode;
@@ -54,7 +53,6 @@ use codex_login::auth::AgentIdentityAuthPolicy;
 use codex_model_provider::create_model_provider;
 use codex_model_provider_info::ModelProviderInfo;
 use codex_model_provider_info::built_in_model_providers;
-use codex_models_manager::bundled_models_response;
 use codex_models_manager::model_info;
 use codex_models_manager::test_support::construct_model_info_offline_for_tests;
 use codex_models_manager::test_support::get_model_offline_for_tests;
@@ -294,11 +292,6 @@ impl StepContext {
 
 mod guardian_tests;
 
-struct InstructionsTestCase {
-    slug: &'static str,
-    expects_apply_patch_description: bool,
-}
-
 fn user_message(text: &str) -> ResponseItem {
     ResponseItem::Message {
         id: None,
@@ -1616,67 +1609,6 @@ async fn user_shell_commands_remain_login_shells_when_model_login_shells_are_dis
     Ok(())
 }
 
-#[tokio::test]
-async fn get_base_instructions_no_user_content() {
-    let prompt_path = codex_utils_cargo_bin::find_resource!(
-        "tests/fixtures/prompt_with_apply_patch_instructions.md"
-    )
-    .expect("resolve prompt fixture path");
-    let prompt_with_apply_patch_instructions =
-        std::fs::read_to_string(prompt_path).expect("read prompt fixture");
-    let models_response = bundled_models_response()
-        .unwrap_or_else(|err| panic!("bundled models.json should parse: {err}"));
-    let model_info_for_slug = |slug: &str, config: &Config| {
-        let model = models_response
-            .models
-            .iter()
-            .find(|candidate| candidate.slug == slug)
-            .cloned()
-            .unwrap_or_else(|| panic!("model slug {slug} is missing from models.json"));
-        model_info::with_config_overrides(model, &config.to_models_manager_config())
-    };
-    let test_cases = vec![
-        InstructionsTestCase {
-            slug: "gpt-5.4",
-            expects_apply_patch_description: false,
-        },
-        InstructionsTestCase {
-            slug: "gpt-5.4-mini",
-            expects_apply_patch_description: false,
-        },
-        InstructionsTestCase {
-            slug: "gpt-5.5",
-            expects_apply_patch_description: false,
-        },
-        InstructionsTestCase {
-            slug: "gpt-5.2",
-            expects_apply_patch_description: false,
-        },
-    ];
-
-    let (session, _turn_context) = make_session_and_context().await;
-    let config = test_config().await;
-
-    for test_case in test_cases {
-        let model_info = model_info_for_slug(test_case.slug, &config);
-        let model_instructions = model_info.get_model_instructions(config.personality);
-        if test_case.expects_apply_patch_description {
-            assert_eq!(
-                model_instructions.as_str(),
-                prompt_with_apply_patch_instructions
-            );
-        }
-
-        {
-            let mut state = session.state.lock().await;
-            state.session_configuration.base_instructions = model_instructions.clone();
-        }
-
-        let base_instructions = session.get_base_instructions().await;
-        assert_eq!(base_instructions.text, model_instructions);
-    }
-}
-
 #[tokio::test]
 async fn reload_user_config_layer_updates_effective_apps_config() {
     let (session, _turn_context) = make_session_and_context().await;
@@ -4876,13 +4808,13 @@ async fn turn_context_with_model_updates_model_fields() {
     ));
     turn_context.current_settings.store(Arc::clone(&current));
     let updated = turn_context
-        .with_model("gpt-5.4".to_string(), &session.services.models_manager)
+        .with_model("gpt-5.5".to_string(), &session.services.models_manager)
         .await;
     let expected_model_info = session
         .services
         .models_manager
         .get_model_info(
-            "gpt-5.4",
+            "gpt-5.5",
             &updated.config.as_ref().to_models_manager_config(),
         )
         .await;
@@ -4923,8 +4855,8 @@ async fn turn_context_with_model_updates_model_fields() {
         &updated.current_settings.load_full(),
         &turn_context.current_settings.load_full()
     ));
-    assert_eq!(updated.config.model.as_deref(), Some("gpt-5.4"));
-    assert_eq!(updated.collaboration_mode().model(), "gpt-5.4");
+    assert_eq!(updated.config.model.as_deref(), Some("gpt-5.5"));
+    assert_eq!(updated.collaboration_mode().model(), "gpt-5.5");
     assert_eq!(updated.model_info().as_ref(), &expected_model_info);
     assert_eq!(
         updated.reasoning_effort(),