Fix build

[yinghai]

How to debug

1. Follow https://circleci.com/docs/ssh-access-jobs/ to get into the host where test failed (e.g. https://app.circleci.com/pipelines/github/pytorch/TensorRT/1461/workflows/877d58af-c765-48df-b6c0-e6328723a7fd/jobs/6968)
2. Repeat the failed command in the circleci host: python3 -c "import torch_tensorrt; torch_tensorrt.dump_build_info()" and see that issue can be reproduced.
3. Repeat with LD_DEBUG=libs python3 -c "import torch_tensorrt; torch_tensorrt.dump_build_info()" and locate the torch libraries at /opt/circleci/.pyenv/versions/3.9.4/lib/python3.9/site-packages/torch/lib/libc10_cuda.so.
4. Do nm -C /opt/circleci/.pyenv/versions/3.9.4/lib/python3.9/site-packages/torch/lib/libc10_cuda.so | grep CUDACachingAllocator |grep T and noticed that it seems weird as it doesn't contain the symbol c10::cuda::CUDACachingAllocator::allocator. It seems that the libs are not from torch 1.14 nightly.
5. Check the torch version python3 -c "import torch; print(torch.__version__)" and it showed 1.13.0+cu117. This is very weird as we are supposed to install 1.14.0+cu116.
6. Check the CI pipeline at noticed that during the Install torch-tensorrt stage, we are actually uninstalling the previously installed torch 1.14

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
torchaudio 0.14.0.dev20221114+cu116 requires torch==1.14.0.dev20221114, but you have torch 1.13.0 which is incompatible.

7. Check py/setup.py and find that we are forcing torch version to be "torch>=1.13.0.dev0,<1.14.0". Fix that.

Type of change

Please delete options that are not relevant and/or add your own.

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes
• I have added the relevant labels to my PR in so that relevant reviewers are notified

[yinghai]

@narendasan //tests/core/conversion/converters:test_einsum failed. Please help check. Thanks.

[frank-wei]

Thanks @yinghai for your time on deep diving in this issue. It helps us a lot on unblocking the 1.3 release!

[frank-wei]

Did you forget to mention in the diagnosis that we also need to link with c10_cuda? :-)

[yinghai]

We probably don't need that. But it doesn't hurt.