Are there plans to add support for training LLM/VLM embedding models (e.g., [GME](https://huggingface.co/Alibaba-NLP/gme-Qwen2-VL-2B-Instruct), [VLM2VEC](https://github.com/TIGER-AI-Lab/VLM2Vec/tree/main), [Qwen3 EMbedding](https://github.com/QwenLM/Qwen3-Embedding)) in the future?