fix no worker node error on multi-node cluster (#1487) (#1513)

Signed-off-by: Yaliang Wu <ylwu@amazon.com>
(cherry picked from commit cea1cd6)

Co-authored-by: Yaliang Wu <ylwu@amazon.com>

plugin/src/main/java/org/opensearch/ml/action/prediction/TransportPredictionTaskAction.java

@@ -121,9 +121,8 @@ public void onFailure(Exception e) {
 
             if (cachedMlModel != null) {
                 modelActionListener.onResponse(cachedMlModel);
-            } else if (modelAccessControlHelper.skipModelAccessControl(user)) {
-                executePredict(mlPredictionTaskRequest, wrappedListener, modelId);
             } else {
+                // For multi-node cluster, the function name is null in cache, so should always get model first.
                 mlModelManager.getModel(modelId, modelActionListener);
             }
         }