Split layer_specs to return Submodules instead of ModuleSpecs

[nschank]

What does this PR do ?

In order to safely refactor Submodules classes, I want to make sure I can easily find everywhere those classes are being referenced. A lot of callers manipulate ModuleSpec objects directly and modify the contained submodules, which makes the operations untyped; this offers methods returning Submodules directly instead so that callers can access the object of a known type instead.

I also wrote a helper method to support propagating the type hints/argument suggestions of the wrapped methods between these, see some context in prior PR: #2834).

Relevant 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.

[yashaswikarnati]

/ok to test 3081683

[ko3n1g]

/ok to test 9c1ae14

[ko3n1g]

I had to update your branch to fix an issue with the CI.

[jaredcasper]

What is the provenance of this code? Want to make sure it didn't come from some other project with its own copyright/license.

[nschank]

Nope, I wrote it myself! The idea of it was inspired by this comment on a previous PR: #2454 (comment) but I wasn't basing it on their implementation, just a vague recollection of the idea haha.

[yashaswikarnati]

/ok to test 85d2216

[ko3n1g]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/21996422034