Add make_palette and make_palette_from_data for palette generation (#210)

Add two public functions for generating categorical color palettes:

- `make_palette(n)`: produce n colors from a source palette, optionally
  reordered for maximum perceptual contrast or colorblind accessibility.
- `make_palette_from_data(sdata, element, color)`: like make_palette but
  derives categories from a SpatialData element and supports spatially-
  aware assignment (spaco-inspired) that maximizes contrast between
  spatially interleaved categories.

Both functions share the same method vocabulary: "default", "contrast",
"colorblind", "protanopia", "deuteranopia", "tritanopia" (non-spatial),
plus "spaco", "spaco_colorblind", "spaco_protanopia", etc. (spatial,
only in make_palette_from_data).

The palette parameter accepts None (scanpy defaults), named palettes
("okabe_ito"), matplotlib colormap names ("tab10"), or explicit color
lists.

Also adds dict[str, str] palette support to render_shapes,
render_points, and render_labels, enabling reuse of generated palettes
across multiple render calls.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: timtreis

src/spatialdata_plot/pl/__init__.py

@@ -1,5 +1,8 @@
+from ._palette import make_palette, make_palette_from_data
 from .basic import PlotAccessor
 
 __all__ = [
     "PlotAccessor",
+    "make_palette",
+    "make_palette_from_data",
 ]