Installing vLLM onto EVO-X2.

Abstract. I download a vLLM Docker image that supports AMD, and the ROCm GPU driver that runs on the Strix Halo platform (AI Max+ 395 APU). I connect the running Docker container to local GGUF AI mode