[quant][fx] Move output share qparam with input ops to backend_config_dict (#75315)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/75315
att, similar to previous PRs, this one moves the ops whose output tensor shares qparams with input
to backend_config_dict
Test Plan:
python test/test_quantization.py TestQuantizeFx
python test/test_quantization.py TestQuantizeFxOps
python test/test_quantization.py TestFXNumericSuiteCoreAPIs
Imported from OSS
Reviewed By: malfet
Differential Revision: D35423791
fbshipit-source-id: b24efc31b79bf6f0c98709a760fb9fba55610c0a
(cherry picked from commit 7535d8da1d052b490566ee60eebce26e68f35ea2)
