[SPARK-51567][ML][CONNECT] Fix DistributedLDAModel.vocabSize

### What changes were proposed in this pull request?
Fix `DistributedLDAModel.vocabSize`

### Why are the changes needed?
```
pyspark.errors.exceptions.connect.SparkException: [CONNECT_ML.ATTRIBUTE_NOT_ALLOWED] Generic Spark Connect ML error. vocabSize in org.apache.spark.ml.clustering.DistributedLDAModel is not allowed to be accessed. SQLSTATE: XX000

JVM stacktrace:
org.apache.spark.sql.connect.ml.MLAttributeNotAllowedException
	at org.apache.spark.sql.connect.ml.MLUtils$.validate(MLUtils.scala:686)
	at org.apache.spark.sql.connect.ml.MLUtils$.invokeMethodAllowed(MLUtils.scala:691)
	at org.apache.spark.sql.connect.ml.AttributeHelper.$anonfun$getAttribute$1(MLHandler.scala:56)
```

### Does this PR introduce _any_ user-facing change?
yes, new api supported

### How was this patch tested?
added test

### Was this patch authored or co-authored using generative AI tooling?
no

Closes apache#50330 from zhengruifeng/ml_connect_lda_vocabSize.

Authored-by: Ruifeng Zheng <ruifengz@apache.org>
Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>

Author: zhengruifeng

python/pyspark/ml/tests/test_clustering.py

@@ -404,6 +404,7 @@ def test_distributed_lda(self):
         self.assertNotIsInstance(model, LocalLDAModel)
         self.assertIsInstance(model, DistributedLDAModel)
         self.assertTrue(model.isDistributed())
+        self.assertEqual(model.vocabSize(), 2)
 
         dc = model.estimatedDocConcentration()
         self.assertTrue(np.allclose(dc.toArray(), [26.0, 26.0], atol=1e-4), dc)

sql/connect/server/src/main/scala/org/apache/spark/sql/connect/ml/MLUtils.scala

@@ -621,8 +621,8 @@ private[ml] object MLUtils {
         "isDistributed",
         "logLikelihood",
         "logPerplexity",
-        "describeTopics")),
-    (classOf[LocalLDAModel], Set("vocabSize")),
+        "describeTopics",
+        "vocabSize")),
     (
       classOf[DistributedLDAModel],
       Set("trainingLogLikelihood", "logPrior", "getCheckpointFiles", "toLocal")),