Add opt-in synchronous Guardian review for native tool calls (#57433)

## What changed

- Add `CoreToolRuntime::requires_guardian_review` to review opted-in function calls before dispatch and stop execution when review rejects the call.
- Deliver proposed calls as native review evidence, preserving encrypted argument metadata. When hooks rewrite arguments, review the updated arguments and discard the original encrypted replay content.
- Deduplicate native call evidence against reviewer history and include paired evidence outputs in input budgets.
- Keep review accounting and metrics for native calls without emitting action-specific client assessment events, and omit approval payloads from debug output.

## Testing

Extend regression coverage for rewritten native calls across review retries, denied calls that never execute their handlers, oversized native evidence, and review metrics without client events.

GitOrigin-RevId: c3b22c0220344581c2b38f233001abaed49cba34

codex-rs/protocol/src/openai_models/guardian.rs

@@ -73,6 +73,7 @@ impl GuardianModelPolicy {
             GuardianScope::Mcp => self.mcp,
             GuardianScope::Network => self.network,
             GuardianScope::Permissions => self.permissions,
+            GuardianScope::NativeTool => Some(GuardianReviewMode::Synchronous),
         }
         .unwrap_or(GuardianReviewMode::Disabled)
     }
@@ -120,6 +121,7 @@ pub enum GuardianScope {
     Mcp,
     Network,
     Permissions,
+    NativeTool,
 }
 
 impl GuardianScope {