[quant][graphmode] Add RemoveReduantDequantize pass (#38434)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/38434
We insert dequantize for each use in order to produce quantization patterns that will
later be fused, after that we should also remove extra dequantize node produced by this operation.
Test Plan: Imported from OSS
Differential Revision: D21597834
fbshipit-source-id: 18dfb2760bbb08932aa4e1d06f96cfc5fb37ed88
