feat(ops): 增强上游错误追踪和新增定时报告服务

- 优化错误日志中间件，即使请求成功也记录上游重试/故障转移事件
- 新增OpsScheduledReportService支持定时报告功能
- 使用Redis分布式锁确保定时任务单实例执行
- 完善依赖注入配置
- 优化前端错误趋势图表展示

backend/internal/repository/ops_repo_dashboard.go

@@ -815,9 +815,9 @@ func (r *opsRepository) queryErrorCounts(ctx context.Context, filter *service.Op
 
 	q := `
 SELECT
-  COALESCE(COUNT(*), 0) AS error_total,
-  COALESCE(COUNT(*) FILTER (WHERE is_business_limited), 0) AS business_limited,
-  COALESCE(COUNT(*) FILTER (WHERE NOT is_business_limited), 0) AS error_sla,
+  COALESCE(COUNT(*) FILTER (WHERE COALESCE(status_code, 0) >= 400), 0) AS error_total,
+  COALESCE(COUNT(*) FILTER (WHERE COALESCE(status_code, 0) >= 400 AND is_business_limited), 0) AS business_limited,
+  COALESCE(COUNT(*) FILTER (WHERE COALESCE(status_code, 0) >= 400 AND NOT is_business_limited), 0) AS error_sla,
   COALESCE(COUNT(*) FILTER (WHERE error_owner = 'provider' AND NOT is_business_limited AND COALESCE(upstream_status_code, status_code, 0) NOT IN (429, 529)), 0) AS upstream_excl,
   COALESCE(COUNT(*) FILTER (WHERE error_owner = 'provider' AND NOT is_business_limited AND COALESCE(upstream_status_code, status_code, 0) = 429), 0) AS upstream_429,
   COALESCE(COUNT(*) FILTER (WHERE error_owner = 'provider' AND NOT is_business_limited AND COALESCE(upstream_status_code, status_code, 0) = 529), 0) AS upstream_529
@@ -870,6 +870,7 @@ error_buckets AS (
   SELECT date_trunc('minute', created_at) AS bucket, COUNT(*) AS cnt
   FROM ops_error_logs
   ` + errorWhere + `
+    AND COALESCE(status_code, 0) >= 400
   GROUP BY 1
 ),
 combined AS (