[quant][graphmode][fx][fp16] Add fp16 support for layer_norm (#52862)
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/52862
Test Plan:
python test/test_quantization.py TestQuantizeFxOps.test_layer_norm
Imported from OSS
Reviewed By: vkuzo
Differential Revision: D26672272
fbshipit-source-id: 4cfdce986efa98db7dc58bf2a62b650e45a69ed0