Re-enable TestNamedTensor.test_big_tensor_repr (pytorch#29407)

Summary:
Pull Request resolved: pytorch#29407

Fixes pytorch#27753.

The bug was that random tensors print subtly differently. This causes
the "names=" tag to appear in slightly different places; sometimes it is
on the same line as the data, sometimes it is on different lines.

For this test, we wanted to know the following:
- printing a big named tensor's repr doesn't crash
- a big named tensor's repr shows the names

This PR changes the test to check those two things.

Test Plan: - run existing tests

Differential Revision: D18428657

Pulled By: zou3519

fbshipit-source-id: 6bcf247ffba010520878a175e766a496028f87d9

Author: zou3519

test/test_namedtensor.py

@@ -275,12 +275,11 @@ def test_no_multiprocessing_support(self):
         with self.assertRaisesRegex(RuntimeError, "NYI"):
             ForkingPickler(buf, pickle.HIGHEST_PROTOCOL).dump(named_tensor)
 
-    @unittest.skip("Issue 27753")
-    def test_big_tensor_repr(self):
+    def test_big_tensor_repr_has_names(self):
         def check_repr(named_tensor):
             unnamed_tensor = named_tensor.rename(None)
-            expected = "{}, names={})".format(repr(unnamed_tensor)[:-1], named_tensor.names)
-            self.assertEqual(repr(named_tensor), expected)
+            names_tag = 'names={}'.format(named_tensor.names)
+            self.assertIn(names_tag, repr(named_tensor))
 
         check_repr(torch.randn(128, 3, 64, 64, names=('N', 'C', 'H', 'W')))