fix: pynvml is pinned when using TRTLLM v13 due to breaking change in 12.0.0

[terrykong]

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Changelog

• Please update the CHANGELOG.md under next version with high level changes in this PR.

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this 

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation? Make sure to also update the NeMo Framework User Guide which contains the tutorials

Checklist when contributing a new algorithm

• Does the trainer resume and restore model state all states?
• Does the trainer support all parallelism techniques(PP, TP, DP)?
• Does the trainer support max_steps=-1 and validation?
• Does the trainer only call APIs defined in alignable_interface.py?
• Does the trainer have proper logging?

Additional Information

• Related to # (issue)

[ko3n1g]

What was the reason for placing this dependency here instead into the requirements.txt?

[terrykong]

What was the reason for placing this dependency here instead into the requirements.txt?

At the moment, we install --no-deps, so it actually wouldn't make a difference, but that's a good point that we can also document this in requirements.txt since we'll eventually rely on that

[ko3n1g]

Thanks for the reminder! SG, I think we’ll anyway be revisiting the dependency management soon

[terrykong]

@ko3n1g Updated requirements.txt + comment

[ko3n1g]

Thanks for this change!