Replace (TF)CommonTestCases for modeling with a mixin.

I suspect the wrapper classes were created in order to prevent the
abstract base class (TF)CommonModelTester from being included in test
discovery and running, because that would fail.

I solved this by replacing the abstract base class with a mixin.

Code changes are just de-indenting and automatic reformattings
performed by black to use the extra line space.

templates/adding_a_new_model/tests/test_modeling_tf_xxx.py

@@ -14,10 +14,12 @@
 # limitations under the License.
 from __future__ import absolute_import, division, print_function
 
+import unittest
+
 from transformers import XxxConfig, is_tf_available
 
 from .test_configuration_common import ConfigTester
-from .test_modeling_tf_common import TFCommonTestCases, ids_tensor
+from .test_modeling_tf_common import TFModelTesterMixin, ids_tensor
 from .utils import CACHE_DIR, require_tf, slow
 
 
@@ -32,7 +34,7 @@ if is_tf_available():
 
 
 @require_tf
-class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
+class TFXxxModelTest(TFModelTesterMixin, unittest.TestCase):
 
     all_model_classes = (
         (

templates/adding_a_new_model/tests/test_modeling_xxx.py

@@ -14,10 +14,12 @@
 # limitations under the License.
 from __future__ import absolute_import, division, print_function
 
+import unittest
+
 from transformers import is_torch_available
 
 from .test_configuration_common import ConfigTester
-from .test_modeling_common import CommonTestCases, ids_tensor
+from .test_modeling_common import ModelTesterMixin, ids_tensor
 from .utils import CACHE_DIR, require_torch, slow, torch_device
 
 
@@ -34,7 +36,7 @@ if is_torch_available():
 
 
 @require_torch
-class XxxModelTest(CommonTestCases.CommonModelTester):
+class XxxModelTest(ModelTesterMixin, unittest.TestCase):
 
     all_model_classes = (
         (XxxModel, XxxForMaskedLM, XxxForQuestionAnswering, XxxForSequenceClassification, XxxForTokenClassification)