Add support for PHI-4 Multimodal

[fire]

Search before asking

• I have searched the Multimodal Maestro issues and found no similar feature requests.

Question

Add support for PHI-4 multimodal as I want to train a combined speech, image and text model.

Additional

No response

[lab176344]

@fire PR is open for it, but I added option to train only the image and text model

[fire]

What is needed to progress? How can I help?

[fire]

I want to train speech to text.