llama.cpp

ggml : fix n_threads_cur initialization with one thread

#9538

Merged