[build] fix: pre-install grpcio-tools before nvidia-resiliency-ext 0.6.0 build

[ko3n1g]

Summary

The automated uv.lock bump in #3081 upgraded nvidia-resiliency-ext from v0.5.0 to 0.6.0. The new version compiles proto files at build time via `grpc_tools.protoc`, but `grpcio-tools` is only declared as a runtime dependency — not in `build-system.requires`. The CI pre-install step doesn't know to install it first, causing:

```
ModuleNotFoundError: No module named 'grpc_tools'
```

Fix: pass `grpcio-tools` via the new `extra-no-build-isolation-packages` input added in NVIDIA-NeMo/FW-CI-templates#440. This pre-installs it before `uv pip install --no-build-isolation -e .` runs, so the build script can find `grpc_tools.protoc`.

Test plan

• CI build-test-publish-wheel / build-wheel passes

🤖 Generated with Claude Code

[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.

[ko3n1g]

/ok to test 116df34

[coderabbitai]

📝 Walkthrough

Walkthrough

The pull request updates a commit reference across two files and extends the build configuration. A commit hash is updated in two locations, and pyproject.toml adds nvidia-resiliency-ext to the build configuration list to disable build isolation for this package due to its proto file compilation requirements.

Changes

Cohort / File(s)	Summary
Commit References .main.commit, 3rdparty/Megatron-LM	Updated commit pointer from 3758b54b2bd9d00f0adbe8bf343a0e6bbcd75d1d to 97e36aaba37245e2fa23d4fda0e956ece21e6c9e in both files, including submodule reference update.
Build Configuration pyproject.toml	Added nvidia-resiliency-ext to [tool.uv] no-build-isolation-package list with explanatory comments regarding proto file compilation via grpc_tools.protoc.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

🚥 Pre-merge checks | ✅ 4

✅ Passed checks (4 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Test Results For Major Changes	✅ Passed	PR contains only minor build configuration changes (6 lines total) with no code logic modifications, new features, or breaking changes. Testing verified through CI build-wheel step.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main change: adding grpcio-tools pre-installation by disabling build isolation for nvidia-resiliency-ext 0.6.0, which is the core fix in the PR.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch ko3n1g/fix/grpcio-tools-build-dep

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In @.main.commit:
- Line 1: The PR mentions adding "nvidia-resiliency-ext" to [tool.uv]
no-build-isolation-package in pyproject.toml but the diff only updates the
commit hash in .main.commit; verify and either add the actual pyproject.toml
change or clarify the relation: update pyproject.toml to include
"nvidia-resiliency-ext" under [tool.uv] no-build-isolation-package (or confirm
the change exists on the commit referenced by .main.commit) and update the PR
description to state whether the .main.commit hash points to the commit that
contains that pyproject.toml modification (reference files: pyproject.toml and
.main.commit).

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: e016e7e9-9932-4c6a-ba10-df9ff0172b3e

📥 Commits

Reviewing files that changed from the base of the PR and between 446baa0 and 116df34.

⛔ Files ignored due to path filters (1)

• uv.lock is excluded by !**/*.lock

📒 Files selected for processing (3)

• .main.commit
• 3rdparty/Megatron-LM
• pyproject.toml

[coderabbitai]

⚠️ Potential issue | 🟠 Major

Inconsistency: PR description mentions pyproject.toml changes, but only commit hash update is provided for review.

The PR objectives state that the fix involves adding "nvidia-resiliency-ext" to [tool.uv] no-build-isolation-package in pyproject.toml to disable build isolation. However, only .main.commit (a commit hash reference) is provided for review.

Please verify:

1. Are the pyproject.toml changes included in this PR?
2. Is the commit hash update in .main.commit related to the build isolation fix, or is it tracking an unrelated upstream dependency update?

Without reviewing the actual pyproject.toml modifications, I cannot confirm that the core fix described in the PR objectives has been implemented.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In @.main.commit at line 1, The PR mentions adding "nvidia-resiliency-ext" to
[tool.uv] no-build-isolation-package in pyproject.toml but the diff only updates
the commit hash in .main.commit; verify and either add the actual pyproject.toml
change or clarify the relation: update pyproject.toml to include
"nvidia-resiliency-ext" under [tool.uv] no-build-isolation-package (or confirm
the change exists on the commit referenced by .main.commit) and update the PR
description to state whether the .main.commit hash points to the commit that
contains that pyproject.toml modification (reference files: pyproject.toml and
.main.commit).

[ko3n1g]

/ok to test c5e847c