[quant] Add 4-bit embedding_bag prepack/unpack support using quint4x2 (#45751)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/45751
Use the torch.quint4x2 dtype to create 4-bit packed tensors
Test Plan:
python test/test_quantization.py TestEmbeddingBagOps
Imported from OSS
Reviewed By: z-a-f
Differential Revision: D24120997
fbshipit-source-id: 6aba2985715a346f6894cf43d5794e104a9ab061
