Support for IBM Power ppc64le architecture.

[RadaKichenin]

Hi,
Thanks for building such a wonderful solution. I am trying to run llmware on IBM Power server (ppc64le) architecture. I was able to install llmware using pip. But I am not able to load and run the gguf models. I have already been working with different gguf models using llama.cpp. It would be great if I can run the llmware optimised gguf models on IBM power server as it has no GPU, but it has a good cpu with an in-core acceleration (Matrix Math Acceleration (MMA)) features. Any pointers in this direction is appreciated.

Thanks & Regards
Radakichenin