Support old and new MCore CUDA graph APIs

[buptzyb]

Summary

Add a small CUDA graph compatibility layer so Megatron-Bridge can work with both the old MCore cuda_graph_scope API and the newer split API using cuda_graph_modules plus cuda_graph_impl="full_iteration". (MCore PR NVIDIA/Megatron-LM#4293)

This updates Bridge call sites that inspect CUDA graph scope directly to go through the compatibility helpers, including full-iteration training/eval paths, local CUDA graph manager creation, delayed wgrad validation, and Qwen3-VL vision CUDA graph config propagation.

Motivation

Megatron-LM is refactoring CUDA graph configuration away from overloaded CudaGraphScope values. Bridge still needs to support existing MCore releases while also remaining compatible with the new API shape.

Validation

• git diff --check
• python -m py_compile src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/transformer_config.py tests/unit_tests/models/qwen_vl/modelling_qwen3_vl/test_qwen3_vl_transformer_config.py

Full pytest was not run locally because dependency setup fails while building nvidia-resiliency-ext without a CUDA installation / CUDA_PATH in this environment.

[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.

[wplf]

/ok to test 440c0b3

[wplf]

/ok to test 6815c35