[Feat] support action forward dynamics finetuning

[WeiChengTseng]

Summary

• Add a launch script for Cosmos3-DROID forward-dynamics post-training.
• Add structured TOML configuration with bf16, distributed sharding, activation checkpointing, and a 20K-step schedule.

Testing

• Not run (configuration and launcher changes only).

[WeiChengTseng]

Update Summary

Addressed the review comments in the latest commit:

• Fixed cookbook-relative paths and anchored them to the launcher location, so the script works regardless of the caller's working directory.
• Removed the dependency on the missing _sft_launcher_common.sh; the launcher is now self-contained with input validation, logging, overrides, and single-/multi-node torchrun setup.
• Added action_fd_droid_posttrain.md with prerequisites, dataset layout, setup instructions, config validation, training commands, and output details.
• Added framework-root resolution and model-config validation so framework-relative assets resolve correctly.

[rickzw]

Not blocking, but since these are not in yet, we have a merge order dependency with cosmos-framework#85.

[rickzw]

@WeiChengTseng can you get NVIDIA/cosmos-framework#85 merged too?

[WeiChengTseng]

I’m waiting for the review of that PR.