feat: 重构 /v1/usage 端点，支持 quota_limited 和 unrestricted 双模式

- quota_limited 模式：返回 Key 级别的总额度、速率限制窗口用量和过期时间
- unrestricted 模式：返回订阅限额或钱包余额信息（向后兼容）
- 新增 model_stats 字段，支持 start_date/end_date 参数查询按模型用量统计
- 提取 buildUsageData/parseUsageDateRange 等辅助方法，减少主函数复杂度
- 新增 APIKeyService.GetRateLimitData 和 UsageService.GetAPIKeyModelStats

backend/internal/service/api_key_service.go

@@ -818,6 +818,11 @@ func (s *APIKeyService) UpdateQuotaUsed(ctx context.Context, apiKeyID int64, cos
 	return nil
 }
 
+// GetRateLimitData returns rate limit usage and window state for an API key.
+func (s *APIKeyService) GetRateLimitData(ctx context.Context, id int64) (*APIKeyRateLimitData, error) {
+	return s.apiKeyRepo.GetRateLimitData(ctx, id)
+}
+
 // UpdateRateLimitUsage atomically increments rate limit usage counters in the DB.
 func (s *APIKeyService) UpdateRateLimitUsage(ctx context.Context, apiKeyID int64, cost float64) error {
 	if cost <= 0 {

backend/internal/service/usage_service.go

@@ -315,6 +315,15 @@ func (s *UsageService) GetUserModelStats(ctx context.Context, userID int64, star
 	return stats, nil
 }
 
+// GetAPIKeyModelStats returns per-model usage stats for a specific API Key.
+func (s *UsageService) GetAPIKeyModelStats(ctx context.Context, apiKeyID int64, startTime, endTime time.Time) ([]usagestats.ModelStat, error) {
+	stats, err := s.usageRepo.GetModelStatsWithFilters(ctx, startTime, endTime, 0, apiKeyID, 0, 0, nil, nil, nil)
+	if err != nil {
+		return nil, fmt.Errorf("get api key model stats: %w", err)
+	}
+	return stats, nil
+}
+
 // GetBatchAPIKeyUsageStats returns today/total actual_cost for given api keys.
 func (s *UsageService) GetBatchAPIKeyUsageStats(ctx context.Context, apiKeyIDs []int64, startTime, endTime time.Time) (map[int64]*usagestats.BatchAPIKeyUsageStats, error) {
 	stats, err := s.usageRepo.GetBatchAPIKeyUsageStats(ctx, apiKeyIDs, startTime, endTime)