A cleaner and more scalable implementation of symbolic tracing (#11763)

Cleaner and more scalable implementation of symbolic tracing with torch.fx, and provides support for new architectures:
- ALBERT
- DistilBERT
- MobileBERT
- MegatronBERT
- GPT2
- GPT Neo

Co-authored-by: Michael Benayoun <michael@huggingface.co>

tests/test_modeling_distilbert.py

@@ -208,6 +208,7 @@ class DistilBertModelTest(ModelTesterMixin, unittest.TestCase):
         if is_torch_available()
         else None
     )
+    fx_ready_model_classes = all_model_classes
     test_pruning = True
     test_torchscript = True
     test_resize_embeddings = True