fix: rename CUDA wheels with tag suffix to avoid upload clash

- all three release builds (pure, cu128, cu130) produce the same .whl filename, causing uploads to overwrite each other
- append the CUDA tag as a local version label (e.g. +cu128, +cu130)

.github/workflows/release.yml

@@ -54,6 +54,9 @@ jobs:
       - name: Build wheel (with CUDA kernels)
         run: |
           CSRC_KERNELS=true pip wheel . --no-deps --no-build-isolation -w dist/
+          for f in dist/*.whl; do
+            mv "$f" "dist/$(basename "$f" .whl)+${{ matrix.cuda_tag }}.whl"
+          done
 
       - uses: actions/upload-artifact@v4
         with: