Track Qwen3.5-related issues

- MCore Path
- Add / track CP support: https://github.com/NVIDIA/Megatron-LM/pull/2645
- Temporaroy solution: https://github.com/NVIDIA-NeMo/RL/pull/2312
- Need to swtich MCore dev branch
- AutoModel Path
- Move the FLA dependency from the dev group ([dependency-groups]) to optional extras ([project.optional-dependencies]) so that NeMo-RL can install it downstream via pkg[extra]. If FLA is not installed, 1) No CP support; 2) Worse performance, related issues/PRs:
- Automodel side: https://github.com/NVIDIA-NeMo/Automodel/pull/1894
- RL side: bump Automodel version
- Fix the default config path where Torch Adam is used without FP32 master weights, as this can slow down convergence.
- TE FusedAdam can be used as a workaround.
- AutoModel should correctly support / apply the FP32 master weight setting. https://github.com/NVIDIA-NeMo/Automodel/pull/1896