Merge branch 'upstream' into concedo_experimental

# Conflicts:
#	examples/server/README.md
#	src/llama-model.cpp

src/llama.cpp

@@ -9471,7 +9471,6 @@ static struct llama_model * llama_model_load_from_file_impl(
         struct llama_model_params params) {
     ggml_time_init();
 
-
     unsigned cur_percentage = 0;
     if (params.progress_callback == NULL) {
         params.progress_callback_user_data = &cur_percentage;