llama.cpp

ce734a8a - ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations (#17977)

ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations (#17977) * feat: implement real Q8_0 * feat: adding cmake option for configuring FP32 quantize group size * typo: set() shall be used --------- Co-authored-by: ngdxzy <zhenyu_xu@uri.edu>