vmin/vmax deprecated kwargs work on 3 render functions but crash on render_images

Description

render_shapes, render_points, and render_labels all accept vmin/vmax via **kwargs and emit a helpful deprecation warning. render_images has a fixed signature with no **kwargs, so the same call crashes with TypeError: got an unexpected keyword argument 'vmin'. Users migrating from the old API face a confusing asymmetry across four otherwise-identical render functions.

Environment

spatialdata-plot: 0.3.4.dev (main, 5cfedc7)
spatialdata: 0.5.0
Python: 3.13

Minimal Reproducible Example

import matplotlib; matplotlib.use("Agg")
import matplotlib.pyplot as plt
import warnings
import dask; dask.config.set({"dataframe.query-planning": False})
import spatialdata as sd
import spatialdata_plot

sdata = sd.datasets.blobs(n_points=50)
img_key = list(sdata.images.keys())[0]
shapes_key = list(sdata.shapes.keys())[0]

# These three work (deprecation warning emitted):
with warnings.catch_warnings():
    warnings.simplefilter("ignore")
    sdata.pl.render_shapes(shapes_key, vmin=0, vmax=1)
    print("render_shapes: OK")

# This crashes:
sdata.pl.render_images(img_key, vmin=0, vmax=200)
print("render_images: OK")

Expected vs. Actual

Expected: render_images emits the same deprecation warning as the other three functions and continues (directing users to use norm=Normalize(vmin=..., vmax=...)).

Actual: TypeError: PlotAccessor.render_images() got an unexpected keyword argument 'vmin'

Fix Sketch

Add **kwargs to render_images and copy the vmin/vmax deprecation guard from the other three functions (basic.py:301-302). Alternatively, explicitly document the asymmetry if it is intentional — though inconsistency across four sibling functions is a usability hazard regardless.

Labels: bug, images, priority: low

Triage tier: Tier 2