BUG: matplotlib 3.9.0 has issue with pandas when using `subplots=True` on `df.plot.scatter()`

[Dr-Irv]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd
df = pd.DataFrame([[5.1, 3.5, 0], [4.9, 3.0, 0], [7.0, 3.2, 1],
                   [6.4, 3.2, 1], [5.9, 3.0, 2]],
                  columns=['length', 'width', 'species'])
df.plot.scatter(x="length", y="width", subplots=True)

Issue Description

With matplotlib 3.9.0 (released 5/15/24), you get:

\lib\site-packages\pandas\plotting\_matplotlib\core.py:896: UserWarning: No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
  ax.legend(loc="best")

Expected Behavior

No warning. It doesn't appear with previous versions of matplotlib

Installed Versions

INSTALLED VERSIONS

commit : d9cdd2e
python : 3.9.16.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.19045
machine : AMD64
processor : Intel64 Family 6 Model 158 Stepping 13, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_United States.1252

pandas : 2.2.2
numpy : 1.26.4
pytz : 2024.1
dateutil : 2.9.0.post0
setuptools : 68.2.2
pip : 22.3.1
Cython : None
pytest : 8.2.0
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : 3.2.0
lxml.etree : 5.2.2
html5lib : 1.1
pymysql : None
psycopg2 : None
jinja2 : 3.1.4
IPython : None
pandas_datareader : None
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : 4.12.3
bottleneck : None
dataframe-api-compat : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.9.0
numba : None
numexpr : 2.10.0
odfpy : None
openpyxl : 3.1.2
pandas_gbq : None
pyarrow : 16.1.0
pyreadstat : 1.2.7
python-calamine : None
pyxlsb : 1.0.10
s3fs : None
scipy : 1.13.0
sqlalchemy : 2.0.30
tables : 3.9.2
tabulate : 0.9.0
xarray : 2024.5.0
xlrd : 2.0.1
zstandard : 0.22.0
tzdata : 2024.1
qtpy : None
pyqt5 : None

[bnavigator]

Also leads to test failure:

[  842s] __________________ TestDataFramePlots.test_plot_scatter_shape __________________
[  842s] [gw0] linux -- Python 3.11.9 /usr/bin/python3.11
[  842s] 
[  842s] self = <pandas.tests.plotting.frame.test_frame.TestDataFramePlots object at 0x7f63b3211590>
[  842s] 
[  842s]     def test_plot_scatter_shape(self):
[  842s]         df = DataFrame(
[  842s]             np.random.default_rng(2).standard_normal((6, 4)),
[  842s]             index=list(string.ascii_letters[:6]),
[  842s]             columns=["x", "y", "z", "four"],
[  842s]         )
[  842s]         # GH 6951
[  842s] >       axes = df.plot(x="x", y="y", kind="scatter", subplots=True)
[  842s] 
[  842s] pandas/tests/plotting/frame/test_frame.py:828: 
[  842s] _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
[  842s] pandas/plotting/_core.py:975: in __call__
[  842s]     return plot_backend.plot(data, x=x, y=y, kind=kind, **kwargs)
[  842s] pandas/plotting/_matplotlib/__init__.py:71: in plot
[  842s]     plot_obj.generate()
[  842s] pandas/plotting/_matplotlib/core.py:503: in generate
[  842s]     self._make_legend()
[  842s] pandas/plotting/_matplotlib/core.py:896: in _make_legend
[  842s]     ax.legend(loc="best")
[  842s] /usr/lib64/python3.11/site-packages/matplotlib/axes/_axes.py:341: in legend
[  842s]     handles, labels, kwargs = mlegend._parse_legend_args([self], *args, **kwargs)
[  842s] /usr/lib64/python3.11/site-packages/matplotlib/legend.py:1364: in _parse_legend_args
[  842s]     _api.warn_external(
[  842s] _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
[  842s] 
[  842s] message = 'No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.'
[  842s] category = None
[  842s] 
[  842s]     def warn_external(message, category=None):
[  842s]         """
[  842s]         `warnings.warn` wrapper that sets *stacklevel* to "outside Matplotlib".
[  842s]     
[  842s]         The original emitter of the warning can be obtained by patching this
[  842s]         function back to `warnings.warn`, i.e. ``_api.warn_external =
[  842s]         warnings.warn`` (or ``functools.partial(warnings.warn, stacklevel=2)``,
[  842s]         etc.).
[  842s]         """
[  842s]         frame = sys._getframe()
[  842s]         for stacklevel in itertools.count(1):
[  842s]             if frame is None:
[  842s]                 # when called in embedded context may hit frame is None
[  842s]                 break
[  842s]             if not re.match(r"\A(matplotlib|mpl_toolkits)(\Z|\.(?!tests\.))",
[  842s]                             # Work around sphinx-gallery not setting __name__.
[  842s]                             frame.f_globals.get("__name__", "")):
[  842s]                 break
[  842s]             frame = frame.f_back
[  842s]         # preemptively break reference cycle between locals and the frame
[  842s]         del frame
[  842s] >       warnings.warn(message, category, stacklevel)
[  842s] E       UserWarning: No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
[  842s]