Fix deprecated string[pyarrow_numpy] in test_to_numpy_pandas_string

[weiji14]

Description of proposed changes

Change from passing in dtype as "string[pyarrow_numpy]" to pd.StringDtype(storage="pyarrow", na_value=np.nan). Fixes TypeError: data type 'string[pyarrow_numpy]' not understood on GMT Dev Tests, e.g. at https://github.com/GenericMappingTools/pygmt/actions/runs/18702196024/job/53333116958#step:15:1167

FutureWarning previously raised, e.g. at https://github.com/GenericMappingTools/pygmt/actions/runs/18514508242/job/52762165871#step:15:1074:

pygmt/tests/test_clib_to_numpy.py::test_to_numpy_pandas_string[string[pyarrow_numpy]]
  /home/runner/work/pygmt/pygmt/pygmt/tests/test_clib_to_numpy.py:385: FutureWarning: The 'pyarrow_numpy' storage option name is deprecated and will be removed in pandas 3.0. Use 'pd.StringDtype(storage="pyarrow", na_value=np.nan)' to construct the same dtype.
  Or enable the 'pd.options.future.infer_string = True' option globally and use the "str" alias as a shorthand notation to specify a dtype (instead of "string[pyarrow_numpy]").
    array = pd.Series(["abc", "defg", "12345"], dtype=dtype)

Xref pandas-dev/pandas#60152

Fixes #

Preview:

Guidelines

• General Guidelines for Pull Request
• Guidelines for Contributing Documentation
• Guidelines for Contributing Code

Slash Commands

You can write slash commands (/command) in the first line of a comment to perform
specific operations. Supported slash command is:

• /format: automatically format and lint the code

[weiji14]

The na_value param was only added in pandas 2.3+ (https://pandas.pydata.org/pandas-docs/version/2.3/reference/api/pandas.StringDtype.html), and we can't drop pandas 2.2 because of SPEC 0 until 2026-01-19. So will need some if-then workaround for now.

[seisman]

Looks good to me.