chore: Bump TRT version to 10.2.0.19

[zewenli98]

Description

Bump to TRT 10.2.0.19

Type of change

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

• New feature (non-breaking change which adds functionality)

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

[zewenli98]

ERROR: [Torch-TensorRT TorchScript Conversion Context] - IBuilder::buildSerializedNetwork: Error Code 6: API Usage Error (Unable to load library: libnvinfer_builder_resource_win.so.10.2.0: libnvinfer_builder_resource_win.so.10.2.0: cannot open shared object file: No such file or directory)

This error occurred on both CI and my local tests. It seems other users are facing the same error:
#2986
NVIDIA/TensorRT#3983

Currently, I noticed there are two solutions to fix the error:

1. downgrade to TRT 10.1
2. pip install tensorrt_llm==0.12.0.dev2024070200

I tried installing tensorrt_llm that works on my local machine.

[HolyWu]

The Python packages have new 10.2.0.post1 released on 2024/7/9. I'm not sure if it solved that error though.

[zewenli98]

The Python packages have new 10.2.0.post1 released on 2024/7/9. I'm not sure if it solved that error though.

@HolyWu Thanks. I just found the issue is about the package tensorrt-cu12-libs. In details,
tensorrt-cu12-libs==10.2.0.post1 works fine.
tensorrt-cu12-libs==10.2.0 results in the error and makes tests crashed totally.
tensorrt-cu12-libs==10.0.1 doesn't cause this error but makes some tests failed.

[HolyWu]

10.3.0 is out 🚀

[zewenli98]

Close this PR in favor of TRT 10.3 upgrade #3071