bfloat16 conv2d and resnet50 tests IREE CPU compiled inference mismatched PyTorch run

[kumardeepakamd]

STR:
git clone https://github.com/nod-ai/SHARK-TestSuite.git
cd e2eshark
(You may need to read the https://github.com/nod-ai/SHARK-TestSuite/blob/main/e2eshark/README.md and make sure you have the right pip install etc.)

Replace the path names for hfhome, -c and -i to point to your Hugging Face Home, Torch MLIR build and IREE builds respectively and run as:

python ./run.py --hfhome /proj/gdba/kumar/HF_HOME -c ../../torch-mlir/build -i ../../mainireee/iree-build --torchtolinalg --todtype bf16 --tests pytorch/operators/conv2d pytorch/models/resnet50

then for each test, cd into cd test-run/pytorch/models/conv2d and test-run/pytorch/models/resnet50 and examine the logs, you can 'cat commands.log' and rerun any step you like. You will see that failedinference.log shows you difference in results

[ScottTodd]

meta: we should find a way to dump reproducers and other artifacts from the e2eshark tests so that people working on these issues can get a quicker start on fixing them.