Add quantized back

[gbeckenkamp]

Quantised models have been skipped for a long time now due to:

# TODO: skipping quantized tests for now due to BC-breaking changes for prepare
# api, enable after PyTorch 1.13 release

but if enabled, those models work. This also improves the coverage of torchbench by allowing some quantised tests to also be validated for real models.

[atalman]

lgtm

[facebook-github-bot]

@atalman has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[JasonMts]

For the record, here are the logs before

pytest benchmark/test.py -k resnet50_quantized_qat --ignore_machine_config

================================================================ test session starts =================================================================
platform linux -- Python 3.10.12, pytest-8.3.4, pluggy-1.5.0
benchmark: 5.1.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /home/user/benchmark
configfile: pyproject.toml
plugins: benchmark-5.1.0, env-1.1.5, timeout-2.3.1
collected 404 items / 404 deselected / 0 selected

============================================================== 404 deselected in 0.20s ===============================================================

and after the fix

pytest benchmark/test.py -k resnet50_quantized_qat --ignore_machine_config

================================================================ test session starts =================================================================
platform linux -- Python 3.10.12, pytest-8.3.4, pluggy-1.5.0
benchmark: 5.1.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /home/user/benchmark
configfile: pyproject.toml
plugins: benchmark-5.1.0, env-1.1.5, timeout-2.3.1
collected 412 items / 408 deselected / 4 selected

benchmark/test.py ....                                                                                                                         [100%]

========================================================= 4 passed, 408 deselected in 34.61s =========================================================