docs: add a Scalebars tutorial

A focused tutorial on drawing physical scalebars via the scalebar_* arguments
of .pl.show(), on the synthetic blobs dataset:

- scalebar_dx (the one required piece) and why it's needed: SpatialData
  coordinate systems carry no unit metadata, so you supply the physical size
  of one axes-unit.
- scalebar_units and automatic magnitude scaling (um -> mm).
- styling via scalebar_params (placement, colour, box, length_fraction,
  scale_loc, fonts) - forwarded verbatim to matplotlib-scalebar.
- fixed_value for an exact bar length.
- cross-links the Visium example for a real-data figure.

Wired into the gallery: card in tutorials/index.md (inside the gallery-cards
markers, root-relative paths) + toctree entry + thumbnail. Addresses part of
scverse/spatialdata-plot#463.

Author: timtreis

_static/img/scalebars.png

@@ -35,6 +35,15 @@ selection for images, and the `datashader` backend for large collections
 of shapes and points.
 :::
 
+:::{grid-item-card} Scalebars
+:link: /notebooks/tutorials/scalebars
+:link-type: doc
+:img-top: /notebooks/_static/img/scalebars.png
+
+Add a physical scalebar with `scalebar_dx`, choose units, and style
+placement, colour, length and fonts through `scalebar_params`.
+:::
+
 ::::
 <!-- gallery-cards-end -->
 
@@ -45,4 +54,5 @@ of shapes and points.
 getting_started
 color_and_palette
 performance
+scalebars
 ```