fix: Pipeline 改为同步初始化 + 日志精简 + 预警去重

1. Pipeline 初始化改回同步（Redis 已正常，不会阻塞）
2. 去掉消息处理时的 10 秒等待循环
3. 飞书长连接启动日志精简（去掉分隔线和重复信息）
4. 性能预警添加 5 分钟冷却期，同类型预警不重复触发

src/core/system_optimizer.py

@@ -198,7 +198,15 @@ class SystemOptimizer:
             logger.error(f"检查安全指标失败: {e}")
     
     def _trigger_performance_alert(self, alert_type: str, message: str):
-        """触发性能预警"""
+        """触发性能预警（同类型预警 5 分钟内不重复）"""
+        now = datetime.now()
+        if not hasattr(self, '_alert_cooldowns'):
+            self._alert_cooldowns = {}
+        last = self._alert_cooldowns.get(alert_type)
+        if last and (now - last).total_seconds() < 300:
+            return  # 冷却中，跳过
+        self._alert_cooldowns[alert_type] = now
+
         try:
             from ..core.models import Alert
             

src/integrations/feishu_longconn_service.py

@@ -55,16 +55,13 @@ class FeishuLongConnService:
         self._executor = ThreadPoolExecutor(max_workers=5, thread_name_prefix="feishu_msg")
         self._msg_count = 0
 
-        # 预初始化 Pipeline（在后台线程，不阻塞启动）
+        # 预初始化 Pipeline
         self._pipeline = None
-        def _init_pipeline():
-            try:
-                self._pipeline = service_manager.get_pipeline()
-                logger.info("✅ MessagePipeline 预初始化成功")
-            except Exception as e:
-                logger.error(f"❌ MessagePipeline 预初始化失败: {e}", exc_info=True)
-        import threading as _threading
-        _threading.Thread(target=_init_pipeline, daemon=True).start()
+        try:
+            self._pipeline = service_manager.get_pipeline()
+            logger.info("✅ MessagePipeline 初始化成功")
+        except Exception as e:
+            logger.error(f"❌ MessagePipeline 初始化失败（将在首条消息时重试）: {e}")
 
     def _handle_message(self, data: P2ImMessageReceiveV1) -> None:
         """收到消息事件，立即提交到线程池，确保 SDK 回调快速返回"""
@@ -179,16 +176,7 @@ class FeishuLongConnService:
             # 使用 Pipeline 统一处理消息
             session_user_id = f"feishu_{chat_id}_{sender_id}"
             try:
-                # 等待 Pipeline 就绪（最多 10 秒）
-                import time as _wait_time
-                for _ in range(100):
-                    if self._pipeline:
-                        break
-                    _wait_time.sleep(0.1)
-                pipeline = self._pipeline
-                if not pipeline:
-                    logger.warning("Pipeline 未就绪，尝试即时初始化")
-                    pipeline = service_manager.get_pipeline()
+                pipeline = self._pipeline or service_manager.get_pipeline()
                 logger.info(f"Pipeline 开始处理: user={session_user_id}")
                 response_data = pipeline.handle_message(
                     user_id=session_user_id,
@@ -275,25 +263,11 @@ class FeishuLongConnService:
             return False
 
     def start(self):
-        """
-        启动长连接客户端
-        这个方法会阻塞当前线程，持续监听飞书事件
-        """
-        logger.info("=" * 80)
-        logger.info(" 启动飞书长连接客户端")
-        logger.info("=" * 80)
-        logger.info(f"📋 配置信息:")
-        logger.info(f"   - App ID: {self.app_id}")
-        logger.info(" 等待消息中... (按 Ctrl+C 停止)")
-        logger.info("=" * 80)
+        """启动长连接客户端（阻塞当前线程）"""
+        logger.info(f"🚀 启动飞书长连接 (App: {self.app_id})")
 
         try:
-            # 创建长连接客户端
             cli = lark.ws.Client(self.app_id, self.app_secret, event_handler=self.event_handler)
-
-            logger.info("正在建立与飞书服务器的连接...")
-
-            # 启动长连接（会阻塞）
             cli.start()
 
         except KeyboardInterrupt: