notebooks: add AdaQuad adaptive quadrature example and normalize checkmark encoding

[MoritzMaibaum]

Motivation

• Provide a demonstrative adaptive quadrature (AdaQuad) example for a 2D spike function inside the project notebook to illustrate certified integration using interval extensions.
• Normalize the pass-message emoji encoding to a Unicode escape to avoid encoding/display issues in the notebook JSON.

Description

• Replaced literal checkmark pass messages ("✅ PASS: ...") with the escaped Unicode sequence "\u2705 PASS: ..." throughout notebooks/test_suite.ipynb to ensure consistent encoding.
• Added a new notebook cell implementing an AdaQuad adaptive partitioning routine: it defines interval_extension_eval, refine_longest_side, adaquad, area/midpoint helpers, computes an adaptive partition, plots a 3D surface with partition midpoints, saves spike_2d_surface_adaquad.pdf, and computes a certified integration interval.
• Minor whitespace/EOF normalization at the end of the notebook file.

Testing

• Executed the notebook sections covering interval-evaluation examples including Random linear network, Zero network, Hand-computable linear network, Identity-style network, and ReLU cases, and all corresponding PASS assertions are present in the notebook outputs.
• The aggregated ReLU notebook tests report passed as shown in the notebook output.
• The newly added AdaQuad cell is present but has execution_count=null and was not executed in this change.

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

intervalNets/notebooks/test_suite.ipynb

Line 911 in 70f02c4

"from scipy import integrate\n",

Add SciPy to declared notebook dependencies

This new cell introduces from scipy import integrate, but SciPy is not declared in pyproject.toml (project or dev extras) or requirements.txt, so a fresh environment following the repo’s install instructions can execute earlier notebook steps and then fail here with ModuleNotFoundError. Because this notebook is the project test suite, the added undeclared import makes the suite non-reproducible unless users manually install an extra package.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".