llama.cpp

ggml-webgpu: Fix compilation error in `ggml_backend_webgpu_debug` in debug mode

#21798

Merged