🪲 StringDtype Compatibility Issue with Pandas 3.0

[e-lo]

Describe the bug

Tests fail on Python 3.11, 3.12, and 3.13, but pass on Python 3.10. The failure occurs during pandera DataFrameModel validation when DataFrames contain pandas StringDtype columns.

TypeError: Cannot interpret '<StringDtype(na_value=nan)>' as a data type

Root Cause

Pandas 2.2+ uses StringDtype for string columns by default
Pandera uses numpy's issubdtype() internally to check dtypes
numpy.issubdtype() cannot handle pandas StringDtype objects, causing the error
This affects Python 3.11+ due to stricter type checking or numpy/pandas version differences

Affected Code

• Any DataFrameModel validation using Series[str] fields
• The TimeStrSeriesSchema using pa.String
  Validation functions in network_wrangler/utils/models.py

Resolution

• Changed pa.String to pandas_engine.NpString() in TimeStrSeriesSchema (uses object dtype, compatible with numpy)
• Added _convert_string_dtype_to_object() function to convert StringDtype columns to object dtype before validation in validate_df_to_model()

Failing tests

• pytest tests/ -v

ails on Python 3.11, 3.12, 3.13; passes on 3.10