Builds docker images to run ik_llama quantized models, llama.cpp compatable models, and vllm backend models.
