Synchronize JdbcSchemaIO provider and Python transform configuration …

…field order (apache#33288)

* Move order of write_batch_size

* Reorder Java schema instead

* Revert jdbc.py changes

.github/trigger_files/beam_PostCommit_Python.json

@@ -1,5 +1,5 @@
 {
   "comment": "Modify this file in a trivial way to cause this test suite to run.",
-  "modification": 5
+  "modification": 6
 }
 

sdks/java/io/jdbc/src/main/java/org/apache/beam/sdk/io/jdbc/JdbcSchemaIOProvider.java

@@ -68,13 +68,13 @@ public Schema configurationSchema() {
         .addNullableField("disableAutoCommit", FieldType.BOOLEAN)
         .addNullableField("outputParallelization", FieldType.BOOLEAN)
         .addNullableField("autosharding", FieldType.BOOLEAN)
-        .addNullableField("writeBatchSize", FieldType.INT64)
         // Partitioning support. If you specify a partition column we will use that instead of
         // readQuery
         .addNullableField("partitionColumn", FieldType.STRING)
         .addNullableField("partitions", FieldType.INT16)
         .addNullableField("maxConnections", FieldType.INT16)
         .addNullableField("driverJars", FieldType.STRING)
+        .addNullableField("writeBatchSize", FieldType.INT64)
         .build();
   }