Rename package to climate_toolkit; full API docs, docs site, and publish metadata

[peetmate]

Summary

Prepares the package for publication. Four areas:

1. Package rename (breaking)

• climate_tookit → climate_toolkit / dist climate-tookit → climate-toolkit, fixing the long-standing typo before first PyPI publication (PyPI names cannot be renamed after publish).
• Mechanical rename across 187 files; console script names were already spelled correctly and are unchanged. Lockfile regenerated.
• Migration: import climate_toolkit; pip install "git+https://github.com/CGIAR-Climate-Data-Hub/climate-toolkit.git".

2. API documentation

• NumPy-style docstrings for all seven public functions (previously 4 had none): every parameter with valid values, return dict structures, Earth Engine auth notes, runnable examples. Package __init__ docstring is now a help(ct) index.
• All facts traced from code, not guessed. Two honest caveats surfaced and documented: evaluate_hazards accepts but does not forward gap_days/min_season_days to the auto season detector; the agera_5 source is backed by the ERA5-Land daily aggregates asset (ECMWF/ERA5_LAND/DAILY_AGGR).

3. Documentation site

• MkDocs Material + mkdocstrings site: getting-started (incl. free GEE registration path), dataset reference page (all 15 sources: resolution, coverage, variables, auth, selection guide), auto-generated API reference, existing workflow docs in nav.
• GitHub Actions workflow deploys to GitHub Pages on push to main (enable Pages → gh-pages branch after first run).

4. Publish metadata & attribution

• CHANGELOG.md, CITATION.cff, docs/homepage/changelog URLs in pyproject.toml, examples/basic_usage.py walkthrough.
• Funding acknowledgements (McKnight CRFS, Alliance Bioversity-CIAT, AIMS Rwanda, CGIAR Climate Data Hub / Climate Action Program AoW1) with logos in README and docs landing page.

Test plan

• pytest: 589 passed, 68 subtests after rename
• mkdocs build --strict clean
• import climate_toolkit + all seven functions resolve; docstrings verified via help()
• Live fetch smoke test (nasa_power, Nairobi 2020)

🤖 Generated with Claude Code

[Sammyjoseph999]

Really solid PR — the rename is thorough (187 files, entry points + dist name updated, test_packaging_metadata fully migrated with no stale refs), the NumPy docstrings are excellent, and I appreciate the honest caveats (the gap_days/min_season_days not being forwarded, and the agera_5 → ECMWF/ERA5_LAND/DAILY_AGGR note). The MkDocs site + dataset reference are a great addition ahead of publication.

One blocking item and a couple of notes:

Must fix — a shipped example is broken by the rename

examples/basic_usage.py still imports the old package name, so it fails with ModuleNotFoundError on import:

• line 83: import climate_tookit as ct
• line 84: from climate_tookit.fetch_data.source_data.sources.utils.models import ClimateVariable
• line 152: print(f"climate_tookit v{ct.__version__}\n")
• line 1 (docstring): Basic usage of climate_tookit ...

All four should be climate_toolkit. Since the whole point here is publication-readiness, a runnable example matters — this looks like the one file the mechanical rename missed.

Minor (non-blocking)

• analysis/package_architecture_summary.md still has ~10 climate_tookit references. Internal doc, not breaking, but worth updating for consistency in a rename PR.

Coordination note

Merging this will break two in-flight PRs that import climate_tookit.* — the API service (#111) and the ERA LTE example (#112). I'll do a rename pass on both once this lands; flagging so it's not missed.

Happy to re-approve as soon as the example imports are fixed.

[peetmate]

Thanks for the careful review — both items fixed in 6029536.

Example imports: confirmed and fixed. Root cause: examples/basic_usage.py was still untracked when the mechanical rename ran over git ls-files, so it was the one file the sweep missed. All four references now climate_toolkit, and the example verified end-to-end (nasa_power fetch returns 366 daily rows; EE-gated demos skip cleanly without credentials).

analysis/package_architecture_summary.md: the ~10 stale references renamed in the same commit.

Re grep -r tookit across tracked files is now clean. Coordination note on #111/#112 understood — thanks for offering to handle the rename pass there.

[Sammyjoseph999]

Both points from my earlier review are addressed in 6029536d — verified on the head branch:

• examples/basic_usage.py now imports climate_toolkit (lines 1, 83, 84, 152).
• analysis/package_architecture_summary.md has no stale references.
• No climate_tookit references remain anywhere except the CHANGELOG entry that documents the rename.

CI is green. Thanks for the quick turnaround — approving. 🚀

Heads-up for after merge: the in-flight #111 (API service) and #112 (ERA LTE example) import climate_tookit.* and will need the rename; I'll push that to both as soon as this lands.