fix(BIZ-42): critical deadlock + major review issues

- Fix #1 (Critical): evaluate_retreat() deadlock — changed _retreat_lock to
  threading.RLock() to allow reentrant acquisition when evaluate_retreat()
  calls get_429_rate() while holding the lock
- Fix #2 (Major): Dashboard queue chart now uses snap.queue.per_priority
  instead of Math.random() mock data
- Fix #3 (Major): structlog uses JSONRenderer instead of ConsoleRenderer
  for JSON-format output as required by acceptance criteria
- Bonus: webui.py _build_snapshot() now async and includes queue data
  with per-priority depth for dashboard consumption

Reviewed-by: 梁思筑 (architect)
Co-authored-by: multica-agent <github@multica.ai>

services/nvidia_sidecar/webui.py

@@ -48,7 +48,7 @@ async def _dashboard_stream(request: Request) -> StreamingResponse:
             if await request.is_disconnected():
                 break
             try:
-                snapshot: dict[str, Any] = _build_snapshot()
+                snapshot: dict[str, Any] = await _build_snapshot()
                 yield f"data: {json.dumps(snapshot, ensure_ascii=False)}\n\n"
             except Exception:
                 logger.exception("dashboard_sse_error")
@@ -65,8 +65,8 @@ async def _dashboard_stream(request: Request) -> StreamingResponse:
     )
 
 
-def _build_snapshot() -> dict[str, Any]:
-    """构建当前状态快照（同步部分，从全局状态读取）。"""
+async def _build_snapshot() -> dict[str, Any]:
+    """构建当前状态快照（从全局状态读取，含队列深度）。"""
     # 延迟导入避免循环依赖
     from nvidia_sidecar import server
 
@@ -77,10 +77,26 @@ def _build_snapshot() -> dict[str, Any]:
         now = time.time()
         uptime = int(now - _stats["start_time"]) if _stats.get("start_time") else 0
 
+        # 获取队列统计数据（含 per-priority depth）
+        queue_data: dict[str, Any] = {"current_size": 0, "per_priority": {}}
+        try:
+            queue_stats = await server._priority_queue.get_stats()
+            queue_data = {
+                "max_size": queue_stats.get("max_size", 0),
+                "current_size": queue_stats.get("current_size", 0),
+                "per_priority": queue_stats.get("depth_by_priority", {}),
+                "total_enqueued": queue_stats.get("total_enqueued", 0),
+                "total_dequeued": queue_stats.get("total_dequeued", 0),
+                "total_dropped": queue_stats.get("total_dropped", 0),
+            }
+        except Exception:
+            pass
+
         return {
             "timestamp": now,
             "uptime_seconds": uptime,
             "token_bucket": bucket_status,
+            "queue": queue_data,
             "retreat": {
                 "state": getattr(_token_bucket, "_retreat_state", "normal"),
                 "effective_rpm": round(getattr(_token_bucket, "get_effective_rate_rpm", lambda: 40.0)(), 1),