Use serving assessments for adaptive Guardian review (#58075)

## What changed

- Match final serving assessments to their response and output item, then retain them with tool invocations and code cells across waits.
- Use matching assessments instead of asynchronous scoring for adaptive review: low-risk actions can skip synchronous review, while elevated-risk actions require it.
- Revalidate captured authorization and policy before approval. Require fresh review for stale or incomplete authorization, preserve explicit review requirements, and discard assessments when hooks rewrite tool input.
- Keep asynchronous scoring as the fallback when an assessment is missing or does not match.

## Testing

Add integration coverage for direct and nested tool calls, hook rewrites, explicit escalation, interrupted responses, catalog-driven policy, incomplete authorization, and assessment retention across code-cell yields and waits.

GitOrigin-RevId: 4346a227204137f018c03c34b302d6f1e9592c96

codex-rs/codex-api/src/guardian_assessment.rs

@@ -4,7 +4,8 @@
 
 use serde_json::Value;
 
-/// A serving-side assessment carried outside model-visible response items.
+/// One final serving assessment emitted immediately before the matching output item is done.
+/// It is carried outside model-visible response items and is not subsequently revised.
 ///
 /// Consumers must validate the response, completed action, and current authorization
 /// before using a low-risk result. This assessment does not grant permission to execute.