fix: add compat with pandas 3

[flying-sheep]

• Closes #
• Tests included or not required because:

• Release notes not necessary because:

TODO:

• Fix test failures caused by hue_order breaks with pandas StringArray mwaskom/seaborn#3893
• Fix test failures caused by fix: add compat with pandas 3 #3929 (comment)

[codecov]

Codecov Report

❌ Patch coverage is 85.29412% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 76.91%. Comparing base (28a1ed4) to head (b565ea9).
✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
src/scanpy/external/exporting.py	0.00%	3 Missing ⚠️
src/scanpy/_utils/__init__.py	80.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3929      +/-   ##
==========================================
+ Coverage   76.89%   76.91%   +0.01%     
==========================================
  Files         117      117              
  Lines       12450    12460      +10     
==========================================
+ Hits         9574     9583       +9     
- Misses       2876     2877       +1     

Flag	Coverage Δ
hatch-test.pre	76.91% <85.29%> (+0.01%)	⬆️

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

Files with missing lines	Coverage Δ
src/scanpy/get/get.py	92.85% <ø> (ø)
src/scanpy/plotting/_anndata.py	85.58% <100.00%> (ø)
src/scanpy/plotting/_dotplot.py	93.48% <100.00%> (ø)
src/scanpy/plotting/_matrixplot.py	96.62% <100.00%> (ø)
src/scanpy/plotting/_scrublet.py	92.15% <100.00%> (ø)
src/scanpy/plotting/_stacked_violin.py	86.06% <100.00%> (ø)
src/scanpy/plotting/_tools/__init__.py	77.02% <100.00%> (+0.25%)	⬆️
src/scanpy/preprocessing/_highly_variable_genes.py	95.03% <100.00%> (ø)
src/scanpy/_utils/__init__.py	77.22% <80.00%> (+0.06%)	⬆️
src/scanpy/external/exporting.py	13.38% <0.00%> (ø)

[flying-sheep]

Other seaborn bug under pandas 3:

sv.parquet.zip

import pandas as pd
import seaborn as sns
from matplotlib import pyplot as plt

df = pd.read_parquet("sv.parquet")

x_axis_order = pd.Index(
    [
        "Gata2_0",
        "Gata1_1",
        "Fog1_2",
        "EKLF_3",
        "Fli1_4",
        "SCL_5",
        "Cebpa_6",
        "Pu.1_7",
        "cJun_8",
        "EgrNab_9",
        "Gfi1_10",
    ],
    dtype="object",
)

sns.violinplot(
    x="genes",
    y="values",
    data=df,
    orient="v",
    hue="genes",
    color=None,
    order=x_axis_order,
    hue_order=x_axis_order,
    cut=0,
    inner=None,
    linewidth=0.2,
    density_norm="width",
)

plt.show(block=True)