Pendulum datetime timezone bug

[vladjohnson]

Describe the bug
Pandera's DataFrameSchema does not properly handle datetime with timezone (datetime64[ns, UTC]) when using Pendulum's datetime. When applying Pandera checks with timezone-aware datetime indexes, it raises a pandera.errors.SchemaError, despite the data type being correctly defined as datetime64[ns, UTC].

Error message:
pandera.errors.SchemaError: expected series 'None' to have type datetime64[ns, UTC], got datetime64[ns, UTC]

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

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

import pandera as pa
import pandas as pd
import pendulum

# Create a sample dataframe
data = {
    "col1": [1.0, 2.0, 3.0],
}
index = pd.DatetimeIndex(
    [pendulum.datetime(2023, 1, 1, tz="UTC"), 
     pendulum.datetime(2023, 1, 2, tz="UTC")]
)

df = pd.DataFrame(data, index=index)

# Define the schema
schema = pa.DataFrameSchema(
    {
        ".*": pa.Column(float, regex=True, nullable=False),
    },
    index=pa.Index(dtype=pd.DatetimeTZDtype(tz="UTC")),
    unique_column_names=True,
)

# Validate the dataframe
schema.validate(df)

Expected behavior

The expected behavior is for the DataFrameSchema validation to pass without errors when using a timezone-aware datetime index (datetime64[ns, UTC]), when utilizing Pendulum's datetime objects.

Desktop (please complete the following information):

• OS: Ubuntu 20.04
• Browser: Chrome
• Version: pandera==0.20.4