Feature: Enable codellama on Intel GPUs

[abhilash1910]

Motivation: Thanks for creating this repository . There is an ongoing effort planned to collaborate from Intel GPU to enable out of the box runtime functionality of code llama on our XPU/GPU Devices. There is also a parallel effort on Llama recipes which is under discussion between Intel and Meta , and we plan to provide consolidated support to all framework/models from Meta research to be able to run on our graphics cards. Mentioning PR : https://github.com/facebookresearch/llama-recipes/pull/116 (llama recipes also in progress).