Back out "[wip][quant][graphmode] produce reference pattern for binary ops and then rewrite to quantized op" (#69713)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69713
Original commit changeset: 456086b308c4
Original Phabricator Diff: D32537714 (https://github.com/pytorch/pytorch/commit/bd8a4a937216433b281a333905175214a12f5fc1)
Reviewed By: jerryzh168
Differential Revision: D32976643
fbshipit-source-id: bea6bf6a2718e42c9efa48a0b0c1dc7fe3893065