ci(cuda): raise Test CUDA job timeout to 480 min

The full CUDA leg (serial pytest ~3h43m + C++ ~20min + LAMMPS) exceeds the
default 360-min GitHub Actions job limit and was cancelled mid-LAMMPS at the
6-hour mark. The self-hosted gpu runner has no hard execution cap, so raise
timeout-minutes to give the suite headroom to finish.

Author: Han Wang

.github/workflows/test_cuda.yml

@@ -17,6 +17,9 @@ jobs:
   test_cuda:
     name: Test Python and C++ on CUDA
     runs-on: gpu
+    # The full CUDA suite (serial pytest ~3h43m + C++ + LAMMPS) exceeds the
+    # default 360-min job limit; the self-hosted GPU runner has no hard cap.
+    timeout-minutes: 480
     # https://github.com/deepmodeling/deepmd-kit/pull/2884#issuecomment-1744216845
     # container:
     #   image: nvidia/cuda:12.9.1-cudnn-devel-ubuntu22.04