mtmd : Support jinja in libmtmd (Only for QwenVL and Qwen Omni)

[alielmorsy]

That code is part of a private repo I’ve been working on. It provides essential support for Jinja in a multi-model setup.
The PR adds two new optional metadata fields for GGUF:

1. tokenizer.ggml.image_token_id:For the image token, if it exists.
2. tokenizer.ggml.audio_token_id: For the audio token, if it exists.

If these tokens do not exist, a fallback is used, similar to the FIM lookup. The current tokens used for images are <|IMAGE|> and <IMAGE>

For the MTMD tokenizer, I maintained backward compatibility and updated the split function to support multiple delimiters, allowing it to work with both the old marker and the preserved tokens.

One final change (only for Qwen models): I removed the image_start and image_end tokens as the model has its own special tokens already.

[alielmorsy]

@ggerganov Could you please check this one?

[ngxson]

IMO this is quite unnecessary complicated. The current system already been able to support image/audio input with jinja templates, without even using image/audio tokens. Why do we need this PR?

[alielmorsy]

@ngxson It simplifies the API to directly convert a dictionary filled with messages to a full prompt the whole magic of the legacy function which surprisely depend on the template under the hood but with some extra layers to clean the messages add the marker,...etc.