chore(beep boop 🤖): Bump uv.lock (main, mcore-dev) (2026-05-19)

[svcnvidia-nemo-ci]

🚀 PR to bump uv.lock in main.

🤖 This PR will be merged automatically once CI passes.

[svcnvidia-nemo-ci]

/ok to test f38df0c

[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.

[yaoyu-33]

MCore bump auto-fix status for dev:

Classification: MCore broke Bridge
Evidence: #3883 still fails UV - Python3.12 - AMD64/Linux - NGC PyTorch job 76718461360 and Launch_Unit_Tests_Core job 76733564272 on 2026-05-19. The logs show ImportError: cannot import name 'InferenceMode' from 'megatron.core.inference.utils' from src/megatron/bridge/inference/vlm/vlm_engine.py:21; the install check also reports missing cache_position docstrings in Qwen3ASR forward methods.
Fix PR: #3889
Guards: added src/megatron/bridge/inference/vlm/_mcore_compat.py with TODO removal condition to remove when Megatron-Core dev exposes InferenceMode from megatron.core.inference.utils; removed stale guards for mtp_block_spec, Mamba hybrid layer helper/kwarg, Muon muon_nesterov, get_mup_config_overrides, async checkpoint/EnergyMonitor imports, and null tokenizer library names.
Validation: local python3 -m py_compile <changed Python files> passed. CW interactive validation on 2026-05-19 passed with srun -A coreai_dlalgo_llm -p interactive ... uv run python -c "import megatron.bridge.inference.vlm.base; import megatron.bridge.inference.vlm.vlm_engine; import megatron.bridge.models.qwen3_asr.hf_qwen3_asr.modeling_qwen3_asr" && uv run python -m pytest tests/unit_tests/inference/vlm/test_base.py tests/unit_tests/inference/vlm/test_vlm_engine.py tests/unit_tests/models/mamba/test_mamba_provider.py tests/unit_tests/recipes/utils/test_optimizer_utils.py tests/unit_tests/training/test_optim.py tests/unit_tests/training/test_state.py tests/unit_tests/training/test_energy_monitor.py tests/unit_tests/training/test_tokenizer.py -q -> 112 passed, 35 warnings. CW interactive uv run pre-commit run --all-files passed. Local uv run could not start because the workstation uv cache path for mamba-ssm==2.3.2.post1 is not writable, so the required hook/test validation was completed in the CW interactive container.
Next action: review fix PR #3889; after it lands, rerun or supersede the current dev bump signal.

[yaoyu-33]

Update: the duplicate fix PR #3889 is closed. The prior fix PR #3876 has been rebased onto current main and no longer carries .dev.commit, 3rdparty/Megatron-LM, or uv.lock changes. Waiting for #3876 to merge before opening any additional dev bump fix PRs.