perf(gateway): 优化负载感知调度

主要改进：
- 优化负载感知调度的准确性和响应速度
- 将 AccountUsageService 的包级缓存改为依赖注入
- 修复 SSE/JSON 转义和 nil 安全问题
- 恢复 Google One 功能兼容性

backend/internal/service/concurrency_service.go

@@ -32,7 +32,6 @@ type ConcurrencyCache interface {
 	// 等待队列计数（只在首次创建时设置 TTL）
 	IncrementWaitCount(ctx context.Context, userID int64, maxWait int) (bool, error)
 	DecrementWaitCount(ctx context.Context, userID int64) error
-	GetTotalWaitCount(ctx context.Context) (int, error)
 
 	// 批量负载查询（只读）
 	GetAccountsLoadBatch(ctx context.Context, accounts []AccountWithConcurrency) (map[int64]*AccountLoadInfo, error)
@@ -201,14 +200,6 @@ func (s *ConcurrencyService) DecrementWaitCount(ctx context.Context, userID int6
 	}
 }
 
-// GetTotalWaitCount returns the total wait queue depth across users.
-func (s *ConcurrencyService) GetTotalWaitCount(ctx context.Context) (int, error) {
-	if s.cache == nil {
-		return 0, nil
-	}
-	return s.cache.GetTotalWaitCount(ctx)
-}
-
 // IncrementAccountWaitCount increments the wait queue counter for an account.
 func (s *ConcurrencyService) IncrementAccountWaitCount(ctx context.Context, accountID int64, maxWait int) (bool, error) {
 	if s.cache == nil {