Feature Request: Support dots.ocr

[congson1293]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

dots.ocr is an OCR model released by rednote that surpasses Gemini 2.5 Pro on benchmarks such as OmniDocBench:
https://huggingface.co/rednote-hilab/dots.ocr

Motivation

Dots.ocr is the great lightweight multilingual VLM model with a score even better than InternVL 1B & 2B. I saw two issues #16161 #15979 requesting support for the model, but they were stale, so I recreate this issue, hoping @ngxson and @CISC can bring the model into llama.cpp.

Possible Implementation

No response

[CISC]

I'll happily review a PR, but I don't think I'll be submitting any more model PRs myself in the foreseeable future (we'd need more reviewer capacity first).