llama.cpp

0137ef88 - ggml : extend `enum ggml_log_level` with `GGML_LOG_LEVEL_DEBUG` (#4579)