llama.cpp

ggml : update ggml_backend_cpu_device_supports_op

#10867

Merged