Build failures since https://github.com/ggerganov/whisper.cpp/pull/2573

[ericcurtin]

containers/ramalama#474

Just tagging @bmahabirbu because Im on holidays and hes one of the folk that looks at the cuda builds.

Such as:

-- Configuring done (6.5s)
CMake Error in ggml/src/ggml-cuda/CMakeLists.txt:
CUDA_ARCHITECTURES is set to "native", but no GPU was detected.

-- Generating done (0.0s)
CMake Generate step failed. Build files cannot be regenerated correctly.
Error: building at STEP "RUN chmod +x /scripts/*.sh && /scripts/build_llama_and_whisper.sh "cuda" "$LLAMA_CPP_SHA" "$WHISPER_CPP_SHA" "/tmp/install" "-DGGML_CUDA=ON" "-DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undefined"": while running runtime: exit status 1

and.....

[ 38%] Building C object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-aarch64.c.o
/whisper.cpp/ggml/src/ggml-kompute/ggml-kompute.cpp:7:10: fatal error: shaderop_scale.h: No such file or directory
7 | #include "shaderop_scale.h"
| ^~~~~~~~~~~~~~~~~~
compilation terminated.
gmake[2]: *** [ggml/src/ggml-kompute/CMakeFiles/ggml-kompute.dir/build.make:76: ggml/src/ggml-kompute/CMakeFiles/ggml-kompute.dir/ggml-kompute.cpp.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:337: ggml/src/ggml-kompute/CMakeFiles/ggml-kompute.dir/all] Error 2
gmake[1]: *** Waiting for unfinished jobs....
[ 41%] Building C object ggml/src/ggml-cpu/CMakeFiles/ggml-cpu.dir/ggml-cpu-quants.c.o
[ 45%] Linking CXX static library libggml-cpu.a
[ 45%] Built target ggml-cpu
gmake: *** [Makefile:136: all] Error 2
Error: building at STEP "RUN chmod +x /scripts/*.sh && /scripts/build_llama_and_whisper.sh "ramalama" "$LLAMA_CPP_SHA" "$WHISPER_CPP_SHA" "/usr" "-DGGML_KOMPUTE=1"": while running runtime: exit status 2