Add Vision Transformer and ViTFeatureExtractor (#10950)

* Squash all commits into one

* Update ViTFeatureExtractor to use image_utils instead of torchvision

* Remove torchvision and add Pillow

* Small docs improvement

* Address most comments by @sgugger

* Fix tests

* Clean up conversion script

* Pooler first draft

* Fix quality

* Improve conversion script

* Make style and quality

* Make fix-copies

* Minor docs improvements

* Should use fix-copies instead of manual handling

* Revert "Should use fix-copies instead of manual handling"

This reverts commit fd4e591bce4496d41406425c82606a8fdaf8a50b.

* Place ViT in alphabetical order

Co-authored-by: Lysandre <lysandre.debut@reseau.eseo.fr>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

tests/test_image_utils.py

@@ -264,7 +264,9 @@ class ImageFeatureExtractionTester(unittest.TestCase):
 
         # During the conversion rescale and channel first will be applied.
         expected = array.transpose(2, 0, 1).astype(np.float32) / 255.0
-        expected = (expected - np.array(mean)[:, None, None]) / np.array(std)[:, None, None]
+        np_mean = np.array(mean).astype(np.float32)[:, None, None]
+        np_std = np.array(std).astype(np.float32)[:, None, None]
+        expected = (expected - np_mean) / np_std
         self.assertTrue(np.array_equal(normalized_image, expected))
 
     def test_normalize_array(self):