fix: harden concurrency limits and high-RPM runtime paths

Bound request, stream, queue, and shutdown resource lifetimes. Reduce scheduler and Redis hot-path work and isolate database maintenance. Include regression coverage, load probes, and concurrency audit results.
2026-10-09 18:59:50 +08:00 · 2026-09-10 08:14:58 +08:00
parent 361952ada9
commit ecc16673eb
149 changed files with 27963 additions and 1926 deletions
@@ -234,7 +234,9 @@ impl Drop for AttemptCancellationGuard {
            );
            return;
        };
+        let usage_producer = state.usage_runtime.track_producer();
        handle.spawn(async move {
+            let _usage_producer = usage_producer;
            settle_cancelled_attempt(state, armed, error_type, error_message).await;
        });
    }