Update CUDA Tile tutorial for CUDA 13.3

[robobryce]

Summary

• update the CUDA Tile tutorial image to the CUDA 13.3.0 devel base
• pin the current CUDA Tile/CUDA Python stack: cuda-tile[tileiras] 1.4.0, cuda-toolkit 13.3.1, cuda-python 13.3.1, cupy-cuda13x 14.1.1, numba 0.66.0, numba-cuda[cu13] 0.30.4, cuda-cccl 1.0.2
• update the notebook docs for CUDA 13.3+ and compute capability 8.x+ GPUs, including A10G
• add a Brev test script that executes the CUDA Tile vector-add notebook

Verification

• python3 brev/test-notebook-format.py tutorials/cuda-tile/notebooks/start.ipynb tutorials/cuda-tile/notebooks/01__cutile_python_intro__vector_add.ipynb
• git diff --check
• docker build -f tutorials/cuda-tile/brev/dockerfile -t ach-cuda-tile:cuda13.3 .
• local container import smoke test for cuda-tile, cuda-python, cuda-toolkit, cupy, numba, and numba-cuda
• Brev A10G g5.xlarge: built commit 1f47992 into ach-cuda-tile:pr-test
• Brev A10G g5.xlarge: docker run --rm --gpus all ... /accelerated-computing-hub/tutorials/cuda-tile/brev/test.bash
  • NVIDIA A10G, driver 595.71.05, CUDA Version 13.3 from inside the container
  • cuda-tile 1.4.0, cuda-python 13.3.1, cupy 14.1.1, numba 0.66.0, numba-cuda 0.30.4
  • notebook executed successfully in 14.083s

Brev cleanup: all A10G Brev instances are deleted; brev ls --json returned {"workspaces": null}.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[github-actions]

👋 Thank you for your contribution!

This pull request is from a forked repository so GitHub Actions will not be able to run CI. A maintainer will review your changes shortly and manually trigger the CI.

@maintainers Please review this PR when you have a chance and follow the instructions in the CONTRIBUTING.md file to trigger the CI.