Pickle breaking changes with minor releases / Column has no attribute default

[Zahlii]

Describe the bug

When using pickle to save schematas, there currently is no guarantee even for minor releases to be compatible. As an example, run the following file twice, once with pandas 0.14.5, once with 0.15.1.

When attempting to unpickle and apply the schema saved with 0.14.5 with 0.15.1, the following error is thrown:

  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/api/pandas/container.py", line 410, in __call__
    return self.validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/api/pandas/container.py", line 341, in validate
    return self._validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/api/pandas/container.py", line 371, in _validate
    return self.get_backend(check_obj).validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/backends/pandas/container.py", line 98, in validate
    results = check(*args)  # type: ignore [operator]
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/backends/pandas/container.py", line 146, in run_schema_component_checks
    result = schema_component.validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/api/pandas/components.py", line 161, in validate
    return self.get_backend(check_obj).validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/backends/pandas/components.py", line 98, in validate
    validate_column(check_obj, column_name)
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/backends/pandas/components.py", line 57, in validate_column
    super(ColumnBackend, self).validate(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/flow-venv-3.9-conda/lib/python3.9/site-packages/pandera/backends/pandas/array.py", line 49, in validate
    if pd.notna(schema.default):
AttributeError: 'Column' object has no attribute 'default'
`
``

- [ x ] I have checked that this issue has not already been reported.
- [ x ] I have confirmed this bug exists on the latest version of pandera.
- [ x ] (optional) I have confirmed this bug exists on the master branch of pandera.


```python

import glob
import pickle as pkl
import pandera as pda
import pandas as pd
x = pd.DataFrame({"A": [1,2,3]})
s = pda.DataFrameSchema({"A": pda.Column(pda.Int, nullable=True)})

with open(f"schema_{pda.__version__}.pkl", "wb") as f:
    pkl.dump(s, f)


for pth in glob.glob("schema*.pkl"):
    print(pth)
    with open(pth, "rb") as f:
        s2 = pkl.load(f)
    s2(x)

Expected behavior

Script runs through with both versions.

Desktop (please complete the following information):

• OS: MacOS
• Version: 0.15.1

[kykyi]

@Zahlii this was an attribute added in Release 0.15, I've opened a small PR to check this attribute exists before calling it. This should fix your issue!

[kykyi]

This can be closed