RPC: Fix Python get_object_from_java missing END_OF_OBJECT on batch boundary

When serialized data items are an exact multiple of the handler's
batchSize, END_OF_OBJECT ends up alone in a separate batch that
receiver.receive() never pulls. Drain the pending batch after
receive() completes — analogous to Java/JS's explicit q.take().

Add integration test with small batchSize to exercise the fix.

Author: knutwannheden

rewrite-python/rewrite/src/rewrite/rpc/server.py

@@ -196,17 +196,21 @@ def pull_batch() -> List[Dict[str, Any]]:
     # Receive and deserialize the object (applies diffs to before state)
     try:
         obj = receiver.receive(before, q)
+
+        # After receive() completes, END_OF_OBJECT may still be pending in a
+        # separate batch (happens when data items are an exact multiple of the
+        # handler's batchSize). Drain it — analogous to Java's explicit
+        # q.take() after receive().
+        if not received_end:
+            pull_batch()
+        if not received_end:
+            raise RuntimeError(f"Did not receive END_OF_OBJECT marker for object {obj_id}")
     except Exception:
         # Reset our tracking of the remote state so the next interaction
         # forces a full object sync (ADD) instead of a delta (CHANGE).
         remote_objects.pop(obj_id, None)
         raise
 
-    # Verify we received the complete object (END_OF_OBJECT was in the final batch)
-    # This matches Java's RewriteRpc.java line 474-475 which explicitly checks for END_OF_OBJECT
-    if not received_end:
-        raise RuntimeError(f"Did not receive END_OF_OBJECT marker for object {obj_id}")
-
     if obj is not None:
         # Update our understanding of what Java has
         remote_objects[obj_id] = obj