Add option for LoRA with Transformer Engine op fuser

[michal2409]

What does this PR do ?

Adding feature from NVIDIA-NeMo/Speech#14411

This PR adds a variant of the LoRALinear module using Transformer Engine's operation fuser. It also adds logic in the existing LoRALinear module so that it can call the fused variant. This is an experimental feature.

Changelog

• Add specific line by line info of high level changes in this PR.

GitHub Actions CI

See the CI sectionin the Contributing doc for how to trigger the CI. A Nvidia developer will need to approve and trigger the CI for external contributors.

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?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

If you haven't finished some of the above items you can still open "Draft" PR.

Additional Information

• Related to # (issue)

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[cuichenx]

LGTM, can you add some tests for the fused path?

[michal2409]

I've added a test for fused path

[ananthsub]

/ok to test 1d0a306

[michal2409]

/ok to test 02cb88a