Fix quantized::linear QuantFusion patterns (#26414)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/26414
Fix the patterns after changes to prepack functions(https://github.com/pytorch/pytorch/pull/25626)
Test Plan:
pytho test/test_jit.py 'TestJit.test_quant_fusion'
Imported from OSS
Differential Revision: D17465553
fbshipit-source-id: 7df6a6aa8389bb4a7a370c65ade4c2585b45b882