Add a knob to throttle the max allowed inflight offload in fine grained offloading

[nanz-nv]

What does this PR do ?

Collaboration with @eee4017 @lhb8125

What

Introduces fine_grained_offloading_max_inflight_offloads on TransformerConfig: for each FGO group name, limits how many completed GPU→CPU offloads can stay unjoined on the default stream before older D2H events are waited on. The same cap applies to every name; each name has its own pending queue. None disables this behavior (unchanged from before for callers who don’t set it).

Why

Under full-iteration CUDA graph capture, the main stream may not otherwise synchronize with D2H work; joining when the per-name depth exceeds the cap gives a tunable backpressure knob and can help with reserved-memory / allocator behavior without a byte-based budget.

Also

Fix the FGO + local CUDA graph validation: compare cuda_graph_scope to [CudaGraphScope.full_iteration] after scope normalization, not the string "full_iteration".
Update the hybrid MoE golden config test snapshot for the new field.

dev PR: #4514

⚠️ For major changes (either in lines of code or in its impact), please make sure to first share a design doc with the team. If you're unsure what's the best way to do so, contact the @mcore-oncall.

Issue tracking

For PRs from open-source community contributors:

• New features: a linked issue is required. Please open a feature request and reference it here before submitting the PR.
• Small updates (bug fixes, minor improvements): a linked issue is recommended and will accelerate the PR review process.

Linked issue:

Contribution process

Pre-checks

• 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

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!

All PRs start as draft. If you open a non-draft PR, it will be automatically converted to draft.

Step 1: Mark PR as "Ready for Review"

1. When your PR is ready, click Ready for Review.
2. An oncall reviewer is auto-assigned and expert reviewers are notified based on your changes.
   • Some PRs may jump straight to step 2. This is determined by .github/CODEOWNERS.

⚠️ Only mark as ready once merge-conflicts are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

Step 2: Final Review

For PRs that change megatron/core, once all expert reviewers have approved, the Final Review label is applied automatically and final reviewers are assigned.

For PRs outside megatron/core, this step is skipped.

Step 3: Approved

Once all required reviewers have approved, the Approved label is applied automatically.

Merge

Any member of mcore-engineers will be able to merge your PR.

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.

[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.

[github-actions]

This PR has been automatically converted to draft because all PRs must start as drafts.

When you are ready for review, click Ready for Review to begin the review process. This will:

1. Add the oncall reviewer (optional reviewer)
2. Add required review teams based on your changes

See the contribution guide for more details.