Add ctis.arange, a centered variant of named_arrays.arange

[roytsmart]

Summary

Adds ctis.arange(start, stop, axis, step), a centered counterpart of named_arrays.arange.

na.arange starts exactly at start and leaves the remainder of the range as a gap on the right. ctis.arange fits as many step-spaced samples as fit in [start, stop] and centers them, splitting the leftover evenly between both ends. This is convenient for building a coordinate grid with a fixed pitch centered on a field of view instead of biased toward one edge.

na.arange(0, 10, axis="x", step=3)     # [0, 3, 6, 9]      (gap on the right)
ctis.arange(0, 10, axis="x", step=3)   # [0.5, 3.5, 6.5, 9.5]  (centered)

Notes

• Built on na.linspace, so it is unit-safe. na.arange wraps numpy.arange, which misbehaves with astropy.units.Quantity step.
• Vector support: when start/stop/step are AbstractVectorArrays (and axis is a matching vector of names), each component is centered independently along its own axis, producing an outer-product grid.
• For divisible ranges both endpoints are included (symmetric); for non-divisible ranges it matches na.arange's sample count, just shifted to center.

Dependency

Requires named-arrays >=2.2 (pin bumped ~=2.0 → ~=2.2): the vector-of-ScalarArrays path relies on linspace accepting a scalar-array num, fixed in named-arrays #194 / released in v2.2.0.

Tests

ctis/_arange_test.py: scalar (divisible + non-divisible), Quantity, and vector cases, the latter parametrized over both Quantity-component and ScalarArray-component vectors. 100% coverage of the module; black/ruff clean; docstring jupyter-execute examples verified.

🤖 Generated with Claude Code

https://claude.ai/code/session_01CtdKmedevWkDab6BWupXqQ

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (54f116f) to head (6360c1e).

Additional details and impacted files

@@            Coverage Diff            @@
##              main       #20   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           21        23    +2     
  Lines          643       697   +54     
=========================================
+ Hits           643       697   +54     

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

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

☔ View full report in Codecov by Harness.
📢 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.