main:删除指标脚本并优化指标记录逻辑

变更内容：
- 删除 `start_metrics.sh` 脚本，精简项目结构，移除不再需要的启动逻辑。
- 优化 HTTP 请求指标记录，新增健康检查端点过滤和路径参数规范化功能。
- 更新文档，添加指标过滤及路径规范化的详细说明。
- 提高 Prometheus 指标的性能和可维护性，避免标签基数爆炸。

src/functional_scaffold/main.py

@@ -53,6 +53,27 @@ async def log_requests(request: Request, call_next):
     return response
 
 
+def normalize_path(path: str) -> str:
+    """
+    规范化路径，将路径参数替换为模板形式
+
+    Args:
+        path: 原始路径
+
+    Returns:
+        规范化后的路径
+
+    Examples:
+        /jobs/a1b2c3d4e5f6 -> /jobs/{job_id}
+        /invoke -> /invoke
+    """
+    # 匹配 /jobs/{任意字符串} 模式
+    if path.startswith("/jobs/") and len(path) > 6:
+        return "/jobs/{job_id}"
+
+    return path
+
+
 # 指标跟踪中间件
 @app.middleware("http")
 async def track_metrics(request: Request, call_next):
@@ -60,8 +81,9 @@ async def track_metrics(request: Request, call_next):
     if not settings.metrics_enabled:
         return await call_next(request)
 
-    # 跳过 /metrics 端点本身，避免循环记录
-    if request.url.path == "/metrics":
+    # 跳过不需要记录指标的端点
+    skip_paths = {"/metrics", "/readyz", "/healthz"}
+    if request.url.path in skip_paths:
         return await call_next(request)
 
     gauge_incr("http_requests_in_progress")
@@ -79,13 +101,15 @@ async def track_metrics(request: Request, call_next):
         raise e
     finally:
         elapsed = time.time() - start_time
+        # 使用规范化后的路径记录指标
+        normalized_path = normalize_path(request.url.path)
         incr(
             "http_requests_total",
-            {"method": request.method, "endpoint": request.url.path, "status": status},
+            {"method": request.method, "endpoint": normalized_path, "status": status},
         )
         observe(
             "http_request_duration_seconds",
-            {"method": request.method, "endpoint": request.url.path},
+            {"method": request.method, "endpoint": normalized_path},
             elapsed,
         )
         gauge_decr("http_requests_in_progress")