feat: summary() returns dict of stats along with printing and added validation tests #315

[archi3103]

Fixes #315

This PR addresses the issue by modifying the summary() method to return a dictionary containing model and term statistics.

Changes

• Updated summary() to create and return stats_dict.
• I made sure to store the actual numbers for things like AIC and EDoF, so they stay accurate if someone wants to use them in a script.
• Updated the function comments to show that it returns a dictionary now.
• Added a new test file pygam/tests/test_summary_return.py to make sure the dictionary is returning the right numbers from the model.

Why this is a good modification

Currently, the summary() method only prints information to the console. But if someone wants to use values like the AIC or p-values in a loop to compare a bunch of different models, they have to copy the numbers by hand.

By having the function return a dictionary, it's much easier to work with the data in code or save it for later. I also made sure to keep the full numbers (not just the rounded ones in the table) so they stay accurate for calculations.

No changes to current functionality
I was careful to make sure this doesn't break anything for existing users. The table still prints in the console exactly as it did before.

If someone doesn't want or needs the dictionary, they can just call gam.summary() like they always have. In Python, if you don't assign the output of a function to a variable, the return value is just ignored, so it won't affect their workflow or clutter their terminal. This makes the change purely additive.

Testing

• Full test suite passed locally (including the new validation test).
• I also checked that the printed table still looks exactly the same as before.