CI: Test against `cuda-cccl`

[leofang]

Follow the style of 3rd party tests that we have:

numba-cuda/.github/workflows/ci-new.yaml

Lines 237 to 272 in 9de460e

	test-thirdparty-cudf:
	needs:
	- build-linux-64
	- compute-matrix
	uses: ./.github/workflows/wheels-test.yaml
	with:
	build_type: pull-request
	script: "ci/test_thirdparty_cudf.sh"
	matrix: ${{ needs.compute-matrix.outputs.TEST_MATRIX }}
	# TODO: Enable for CUDA 13 when a supporting version of cuDF is available
	matrix_filter: map(select(.ARCH == "amd64" and (.CUDA_VER | split(".") | .[0] | tonumber == 12))) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
	test-thirdparty-nvmath:
	if: ${{ github.ref_name == 'main' }}
	needs:
	- build-linux-64
	- compute-matrix
	uses: ./.github/workflows/wheels-test.yaml
	with:
	build_type: pull-request
	script: "ci/test_thirdparty_nvmath.sh"
	matrix: ${{ needs.compute-matrix.outputs.TEST_MATRIX }}
	# TODO: Enable for CUDA 13 when a supporting version of nvmath-python is available
	matrix_filter: map(select(.ARCH == "amd64" and (.CUDA_VER | split(".") | .[0] | tonumber == 12))) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
	test-thirdparty-awkward:
	needs:
	- build-linux-64
	- compute-matrix
	uses: ./.github/workflows/wheels-test.yaml
	with:
	build_type: pull-request
	script: "ci/test_thirdparty_awkward.sh"
	matrix: ${{ needs.compute-matrix.outputs.TEST_MATRIX }}
	# TODO: Enable for CUDA 13 in future
	matrix_filter: map(select(.ARCH == "amd64" and (.CUDA_VER | split(".") | .[0] | tonumber == 12))) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))