Update quality tooling for formatting (#21480)

* Result of black 23.1

* Update target to Python 3.7

* Switch flake8 to ruff

* Configure isort

* Configure isort

* Apply isort with line limit

* Put the right black version

* adapt black in check copies

* Fix copies

tests/models/groupvit/test_modeling_tf_groupvit.py

@@ -23,8 +23,8 @@ import unittest
 from importlib import import_module
 
 import numpy as np
-
 import requests
+
 from transformers import GroupViTConfig, GroupViTTextConfig, GroupViTVisionConfig
 from transformers.testing_utils import (
     is_pt_tf_cross_test,
@@ -96,7 +96,6 @@ class TFGroupViTVisionModelTester:
         self.seq_length = num_patches
 
     def prepare_config_and_inputs(self):
-
         rng = random.Random(0)
         pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size], rng=rng)
         config = self.get_config()
@@ -452,7 +451,6 @@ class TFGroupViTTextModelTester:
 
 @require_tf
 class TFGroupViTTextModelTest(TFModelTesterMixin, unittest.TestCase):
-
     all_model_classes = (TFGroupViTTextModel,) if is_tf_available() else ()
     test_pruning = False
     test_head_masking = False