uv: set minimum numpy version to 2.0 (#810)

[SaguaroDev]

Closes #810.

Running the suite on numpy 1.x breaks the unit tests, so the floor should be numpy 2.x. The existing dependency set already implies a 2.x-compatible window — numba caps numpy at <2.4, and pandas>=2.3.3 needs numpy>=1.26 on py3.12+ — but nothing pinned a 2.x floor explicitly. This sets it.

Change: "numpy" → "numpy>=2.0" in [project.dependencies].

Why 2.0 and not higher: 2.0 is the lowest numpy 2.x, it satisfies every transitive constraint in the dep tree (pandas, scikit-learn, sparse, numba, matplotlib all resolve against it), and requires-python>=3.10 is compatible with numpy 2.0+. Going higher would over-constrain without cause.

Verification (the "uv solves AND passes tests" check from the issue): I built an isolated env and pinned numpy to the floor (numpy==2.0.2). uv resolved the full environment cleanly, and at that floor the suite passes:

716 passed, 1 skipped, 12 xfailed in 133.41s

(At the floor the resolver also pulled pandas 3.0.3 / sparse 0.18 / numba 0.65.1, so the floor holds even against the newest pandas — not just the currently-pinned 2.3.3.)

I left uv.lock out of the diff so it regenerates in CI; happy to commit a refreshed lock if you'd prefer it in the PR.

---

Note

Low Risk
Packaging-only constraint change; no runtime or library code is modified.

Overview
Pins NumPy 2.x as the minimum runtime dependency by changing numpy to numpy>=2.0 in [project.dependencies] in pyproject.toml.

This aligns declared requirements with the stack the project already expects (tests fail on NumPy 1.x) without changing application code.

^{Reviewed by Cursor Bugbot for commit a822acc. Bugbot is set up for automated code reviews on this repo. Configure here.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.90%. Comparing base (8a9a46e) to head (a822acc).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #899   +/-   ##
=======================================
  Coverage   86.90%   86.90%           
=======================================
  Files          87       87           
  Lines        4932     4932           
  Branches      624      624           
=======================================
  Hits         4286     4286           
  Misses        456      456           
  Partials      190      190           

Flag	Coverage Δ
unittests	86.90% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.