Tutorials/CUDA Tile: Add cuTile Python notebooks for matrix add, transpose, and activation functions

[rbinyahib]

Summary

• Add three new cuTile Python tutorial notebooks: matrix add (02), transpose (03), and activation functions (04)
• Improve notebook 01 with a tile-vs-SIMT explanation and validation print statements
• Update README.md and start.ipynb to include the new notebook entries

Notebooks

01 - Vector Add (updated): Added an inline explanation of the tile-vs-SIMT programming model shift and print statements after each validation assertion.

02 - Matrix Add (new): Introduces 2D tiles and shows how the grid covers a matrix with both row and column dimensions.

03 - Transpose (new): Covers transposing a tile with ct.transpose() and swapping the store index. Includes a symmetric sum exercise with an inline solution.

04 - Activation Functions (new): Covers ct.maximum and ct.minimum as named cuTile operations via ReLU and clamp kernels. Includes a clamp exercise with an inline solution.

Notes

The pre-push hook reports format failures on three pre-existing notebooks that I did not modify:

• events/gtc_spring_2025/talks/cuda_techniques_compute_and_instruction_S72685/line_intersection/jupyter/line_intersection.ipynb
• events/pycon_2025/ai_with_a_conscience.ipynb
• docs/notebook_template.ipynb

These failures exist on main independently of this PR.

[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.

[rbinyahib]

@brycelelbach @thschmid-nv