Add pre-fitted atomic Gaussian parameter loader and data

[Ao-chuba]

2nd PR for Improvement of the robustness of poisson equation

This pull request implements a zero dependency loader load_atomic_gaussian_params(element) in src/grid/coulomb.py to retrieve pre-fitted $s$-type Gaussian parameters (coefficients and exponents) for elements H, C, N, O, and Cl from a package-safe JSON dataset (atomic_gauss_params.json). The loader supports case insensitive, space stripped element symbols (e.g. "cl", " C ") as well as atomic numbers (e.g. 17), utilizing a module-level lazy cache to avoid repeated disk reads. The packaging configuration in MANIFEST.in has been updated to bundle the dataset, the loader has been exported at the package root level, and unit tests have been added to verify type handling, lookup correctness, and parameter validity.

related to issue -> here

[Copilot]

Pull request overview

This PR adds a package-safe, zero-extra-dependency loader to fetch pre-fitted s-type atomic Gaussian parameters (coefficients/exponents) for a small set of elements (H, C, N, O, Cl), backed by a JSON dataset shipped with the package. This supports the Poisson-solver robustness effort by enabling “promolecular/fit-based” strategies without external fit libraries.

Changes:

• Added load_atomic_gaussian_params(element) with symbol/atomic-number support and lazy JSON caching.
• Added src/grid/data/atomic_gauss_params.json containing the fitted parameters.
• Exported the loader and added unit tests validating basic lookup/shape/value constraints.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/grid/coulomb.py	Implements the JSON-backed atomic parameter loader and exports it.
src/grid/data/atomic_gauss_params.json	Adds the pre-fitted parameter dataset for supported elements.
src/grid/tests/test_coulomb.py	Adds tests for the new loader.
src/grid/__init__.py	Re-exports Coulomb module symbols at package root.
MANIFEST.in	Attempts to include the new JSON dataset in packaged artifacts.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[marco-2023]

These are weak tests, but for now, fine

[Ao-chuba]

RIght i see, i think to improve the tc , we can add in normalization checks or potential accuracy checks ? if that makes sense we can add in those test cases in the next commit or maybe even a different pr.

[marco-2023]

Please fix the comments, also check #317 for the Manifest.in structure

[Ao-chuba]

@marco-2023 thanks for the review, i have updated the pr as requested. please do let me know if any other is required.