Typehint

[zhengp0]

Summary

Add type hints to the public functions and methods in core.py and
utils.py, and fix option_to_list so the per-dimension *_list
options actually work as documented. Also modernize the CI workflow,
which is required because the new typing syntax needs Python >= 3.10.

Typing conventions used:

• numpy.typing.NDArray for real arrays, numpy.typing.ArrayLike for
  the knot inputs that get normalized via np.array/np.asarray
  (the XSpline / NDXSpline constructors).
• float | NDArray (simple unions, no @overload) for the
  scalar-or-array math functions, matching how the code branches on
  np.isscalar.

What changed

• src/xspline/utils.py — annotated all functions (args + returns).
  Notable: order_to_index(shape) is tuple[int, ...] | NDArray
  (tuple in tests, ndarray in core.py); funcs is
  list[Callable[..., float | NDArray]]; outer_flatten(*args: NDArray).
• src/xspline/utils.py (option_to_list) — behavior fix.
  Previously any non-empty list was truthy and collapsed to all-True,
  so per-dimension lists were silently ignored. It now accepts a single
  bool, None, or a real list[bool] (length-checked), returning the
  list as-is. Signature: bool | list[bool] | None.
• src/xspline/core.py — annotated the four bspline_* functions,
  the XSpline / NDXSpline methods, constructors (-> None), and the
  nested integration piece helpers in ifun. The *_list /
  *_extra_list params are typed bool | list[bool] | None to keep the
  scalar/None convenience while enabling real lists.
• tests/test_utils.py — replaced the previously vacuous
  test_utils_option_to_list (its ~(a ^ b) assertion was always
  truthy) with assertions that actually check the values, plus new tests
  for the real-list path and the length-mismatch guard.
• .github/workflows/python-build.yml — bumped the test Python to
  3.11 (the new X | Y / list[bool] syntax needs >= 3.10, which was
  the CI failure), and updated action versions:
  actions/checkout@v2 -> v4, actions/setup-python@v2 -> v5,
  pypa/gh-action-pypi-publish@master -> @release/v1.

[codacy-production]

Not up to standards ⛔

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}