llama.cpp

Feature: Integrate with unified SYCL backend for Intel GPUs

#2690

Merged