[ML] fixing bwc for trained model deployment inference (#88289) (#88315)

When the internal objects were renamed, an inappropriate bwc version restriction was put in place.

This commit fixes this by allowing trained model assignment metadata updates to be serialized to nodes > 8.0.0.

This is OK as the object serialization handles its BWC conditions when serializing over the wire.

This closes: #87959

Author: benwtrent

x-pack/plugin/ml/src/main/java/org/elasticsearch/xpack/ml/inference/assignment/TrainedModelAssignmentMetadata.java

@@ -127,7 +127,7 @@ public String getWriteableName() {
 
     @Override
     public Version getMinimalSupportedVersion() {
-        return Version.V_8_3_0;
+        return Version.V_8_0_0;
     }
 
     @Override
@@ -265,7 +265,7 @@ public String getWriteableName() {
 
         @Override
         public Version getMinimalSupportedVersion() {
-            return Version.V_8_3_0;
+            return Version.V_8_0_0;
         }
 
         @Override

x-pack/qa/rolling-upgrade/src/test/java/org/elasticsearch/upgrades/MLModelDeploymentsUpgradeIT.java

@@ -79,8 +79,10 @@ public void testTrainedModelDeployment() throws Exception {
                     request.addParameter("timeout", "70s");
                 }));
                 waitForDeploymentStarted(modelId);
-                assertInfer(modelId);
-                assertInfer(modelId);
+                // attempt inference on new and old nodes multiple times
+                for (int i = 0; i < 10; i++) {
+                    assertInfer(modelId);
+                }
             }
             case UPGRADED -> {
                 ensureHealth(".ml-inference-*,.ml-config*", (request -> {