llama.cpp

ggml: update SCHED_DEBUG output to use ggml_op_desc()

#22825

Merged