Add support for export ComfyUI compatible checkpoint for diffusion model(e.g., LTX-2)

[ynankani]

What does this PR do

Add support for export ComfyUI compatible checkpoint for diffusion model(e.g., LTX-2)

Type of change:

Overview:
Add support for export ComfyUI compatible checkpoint for diffusion model(e.g., LTX-2)

1. Added a a parameter for merging the base vae, vocoder, connectors in the quantized checkpoint
2. storing quantization metadata and export tool as modelopt , required for ComfyUI compatibility.
3. Internally updating the transformer block prefixes to match the expectation of ComfyUI

Usage

    export_hf_checkpoint(
        pipeline,
        export_dir=EXPORT_DIR,
        merged_base_safetensor_path=BASE_CKPT,  # merge VAE/vocoder from base
    )

Testing

1. Tested with ltx-2 model
   a) initializing a twoStagePipeline object
   b) calling mtq.quantize on transformer with NVFP4_DEFAULT_CFG
   c) then exporting with export_hf_checkpoint passing the param merged_base_safetensor_path to generate merged
   checkpoint
2. Ran the generated checkpoint with step1 on ComfyUI to validate
3. Ran step1 without merged_base_safetensor_path to check backward compatibility.

Before your PR is "Ready for review"

• Make sure you read and follow Contributor guidelines and your commits are signed.
• Is this change backward compatible?: Yes
• Did you write any new necessary tests?: NA
• Did you add or update any necessary documentation?: NA
• Did you update Changelog?: NA

Additional Information

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 73.10%. Comparing base (7c4c9fd) to head (69107c0).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #911   +/-   ##
=======================================
  Coverage   73.10%   73.10%           
=======================================
  Files         205      205           
  Lines       22281    22281           
=======================================
  Hits        16288    16288           
  Misses       5993     5993           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[jingyu-ml]

Left some comments, overall it looks good to me.

[jingyu-ml]

For now, this seems to work only for LTX2.

Are these mapping relationships hard-coded? If so, we should move this logic into a model-dependent function, for example:

model_type = LTX2
merge_function[LTX2](...)

[jingyu-ml]

and this function needs to be moved to https://github.com/NVIDIA/Model-Optimizer/blob/main/modelopt/torch/export/diffusers_utils.py

[jingyu-ml]

As noted above, this function should be model-dependent

merge_function[model_type](...)

[jingyu-ml]

should add more checks to make it more safer
if hf_quant_config is not None and merged_base_safetensor_path is not None: