Migrate MoeLayer submodules from ModuleSpec to Protocols

[nschank]

What does this PR do ?

Defines Protocols representing Experts and SharedExperts, and uses them instead of ModuleSpec to enable typechecking of MoeLayer construction.

I also updated Backend to just return an instance of ExpertsBuilder because that seems to be the real intended use here; I can add a more introspective intermediate step if desired though, it just wasn't needed in any existing use-case.

Associated design doc: Typed ModuleSpec.pdf

Contribution process

flowchart LR
    A[Pre-checks] --> B[PR Tests]
    subgraph Code Review/Approval
        C1[Expert Review] --> C2[Final Review]
    end
    B --> C1
    C2 --> D[Merge]

Loading

Pre-checks

• I want this PR in a versioned release and have added the appropriate Milestone (e.g., Core 0.8)
• I have added relevant unit tests
• I have added relevant functional tests
• I have added proper typing to my code Typing guidelines
• I have added relevant documentation
• I have run the autoformatter.sh on my PR

Code review

The following process is enforced via the CODEOWNERS file for changes into megatron/core. For changes outside of megatron/core, it is up to the PR author whether or not to tag the Final Reviewer team.

For MRs into `main` branch

Feel free to message or comment the @mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!

(Step 1): Add PR label Expert Review

(Step 2): Collect the expert reviewers reviews

1. Attach the Expert Review label when your PR is ready for review.
2. GitHub auto-assigns expert reviewers based on your changes. They will get notified and pick up your PR soon.

⚠️ Only proceed to the next step once all reviewers have approved, merge-conflict are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

(Step 3): Final Review

1. Add Final Review label
2. GitHub auto-assigns final reviewers based on your changes. They will get notified and pick up your PR soon.

(Optional Step 4): Cherry-pick into release branch

If this PR also needs to be merged into core_r* release branches, after this PR has been merged, select Cherry-pick to open a new PR into the release branch.

For MRs into `dev` branch

The proposed review process for `dev` branch is under active discussion.

MRs are mergable after one approval by either eharper@nvidia.com or zijiey@nvidia.com.

Merging your PR

Any member of core-adlr and core-nemo will be able to merge your PR.

[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.

[Phlip79]

/ok to test 8d80937

[nschank]

This kind of test failure is what I had in mind when sending #3425!

I have a generic introspection tool meant to help with this, let me pull that into this PR - will update later today

[nschank]

Updated - a simple get_submodules method which can introspect into both partial and ModuleSpec normally. (I will add get_param and the like as needed)

[nschank]

Resynced after coming back from travel, sorry for delay!

[nschank]

Rebased to fix merge conflict!

[yashaswikarnati]

/ok to test c456795

[yashaswikarnati]

/ok to test 5ae2adb

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/23512655178