fix: Query node panic during sending rpc to worker (#36975)

issue: #36976

Signed-off-by: Wei Liu <wei.liu@zilliz.com>

internal/querynodev2/cluster/worker.go

@@ -209,6 +209,7 @@ func (w *remoteWorker) Stop() {
 		for _, client := range w.clients {
 			client.Close()
 		}
+		return
 	}
 	if err := w.client.Close(); err != nil {
 		log.Warn("failed to call Close via grpc worker", zap.Error(err))