Adjust cpp torch trt logging level with compiler option

[keehyuna]

Description

Log level of python and native TRT logger is set differently. Same level from compiler option should be helpful to work with development version of torch-TRT module.

• python trt logger: when compiler option provide debug=True
• c++ trt logger: compiled at build time

TorchTRTLogger.cpp
#ifndef NDEBUG
static TorchTRTLogger global_logger("[Torch-TensorRT - Debug Build] - ", LogLevel::kDEBUG, true);
#else
static TorchTRTLogger global_logger("[Torch-TensorRT] - ", LogLevel::kWARNING, false);
#endif

And profile option for c++ trt runtime module is enabled by debug level logging. This is helpful to measure/evaluate
performance with development version of torch-TRT module

Fixes # (issue)

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

[narendasan]

What happens in the case that it is a python only build? You might want to use enabled features to gate access to this code to only when the C++ runtime is available.

[keehyuna]

You're right, I see the problem. Updated the change.

AttributeError: '_OpNamespace' 'tensorrt' object has no attribute 'set_logging_level'

[narendasan]

LGTM