fix(fish-speech): use CUDA toolkit ptxas

Prefer an explicitly configured Triton assembler, otherwise use the executable ptxas from CUDA_HOME so torch.compile can target GPU architectures newer than Triton bundled tooling.

Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

backend/python/fish-speech/run.sh

@@ -10,4 +10,9 @@ fi
 # stale when the backend is relocated under /backends at install time.
 export PYTHONPATH="${EDIR}/fish-speech-src${PYTHONPATH:+:${PYTHONPATH}}"
 
+cuda_home=${CUDA_HOME:-/usr/local/cuda}
+if [ -z "${TRITON_PTXAS_PATH:-}" ] && [ -x "$cuda_home/bin/ptxas" ]; then
+    export TRITON_PTXAS_PATH="$cuda_home/bin/ptxas"
+fi
+
 startBackend "$@"