GitLab

Add native support for converting Qwen2 family models (including Qwen2.5)
from safetensors to gguf format so we can run it.