[codex] Add internal auto-compaction opt-out (#28260)

## Summary

- add a default-on `auto_compaction` feature flag as an internal escape
hatch
- skip pre-turn, model-switch/hash, and mid-turn automatic compaction
when the flag is disabled
- preserve manual `/compact` behavior and surface the existing
context-window error when the provider runs out of room
- add integration coverage for disabled pre-turn and mid-turn compaction

## Motivation

Long-running SPO optimization rollouts need the option to preserve their
full context and fail on context exhaustion instead of entering another
compaction window. This deliberately uses the existing feature-flag
mechanism rather than adding a dedicated public config or app-server
API.

Disable it with:

```sh
codex --disable auto_compaction
```

## Testing

- `just test -p codex-features` — 51 passed
- `just test -p codex-core auto_compaction_feature_disabled` — 2 passed
- `just fix -p codex-core -p codex-features`
- `just write-config-schema`
- `just test -p codex-core` — the new compaction tests passed; the
overall local run had 54 unrelated environment failures, primarily
missing first-party test binaries and shell-snapshot timeouts

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

@@ -333,7 +333,10 @@ pub(crate) async fn run_turn(
                 }
 
                 // as long as compaction works well in getting us way below the token limit, we shouldn't worry about being in an infinite loop.
-                if token_limit_reached && needs_follow_up {
+                if turn_context.features.enabled(Feature::AutoCompaction)
+                    && token_limit_reached
+                    && needs_follow_up
+                {
                     if let Err(err) = run_auto_compact(
                         &sess,
                         &turn_context,
@@ -844,6 +847,10 @@ async fn run_pre_sampling_compact(
     turn_context: &Arc<TurnContext>,
     client_session: &mut ModelClientSession,
 ) -> CodexResult<()> {
+    if !turn_context.features.enabled(Feature::AutoCompaction) {
+        return Ok(());
+    }
+
     maybe_run_previous_model_inline_compact(sess, turn_context, client_session).await?;
     let token_status = auto_compact_token_status(sess.as_ref(), turn_context.as_ref()).await;
     // Compact if the configured auto-compaction budget or usable context window is exhausted.

codex-rs/core/src/tools/spec_plan.rs

@@ -708,7 +708,9 @@ fn add_core_utility_tools(context: &CoreToolPlanContext<'_>, planned_tools: &mut
     }
 
     if features.enabled(Feature::TokenBudget) {
-        planned_tools.add_with_exposure(NewContextWindowHandler, ToolExposure::DirectModelOnly);
+        if features.enabled(Feature::AutoCompaction) {
+            planned_tools.add_with_exposure(NewContextWindowHandler, ToolExposure::DirectModelOnly);
+        }
         planned_tools.add(GetContextRemainingHandler);
     }