feat: convert tidy data chapter from pandas to Polars

[meshackamadi]

Summary

This PR converts the Tidy Data chapter from pandas to Polars and updates the accompanying explanations to align with the Polars API and data model.

Changes

• Replaced pandas examples with their Polars equivalents.
• Updated reshaping examples to use unpivot() instead of melt().
• Updated pivot() examples to use the Polars API (on= instead of columns=).
• Replaced pd.read_parquet() with pl.read_parquet().
• Replaced DataFrame creation examples with pl.DataFrame.
• Updated text throughout the chapter to reference Polars instead of pandas.
• Removed pandas index-based operations where Polars has no equivalent.
• Revised explanations to reflect Polars' expression-based workflow.

Documentation updates

• Updated prerequisites to use Polars.
• Renamed the reshaping tools section to reference Polars.
• Improved explanations of tidy data using Polars terminology.

Notes

Testing

• All notebook cells execute successfully.
• Quarto renders successfully.
• pre-commit run --all-files passes.

[meshackamadi]

Changes

Code Migration

• pd.date_range() → pl.date_range() with eager collection
• .pivot(columns=...) → .pivot(on=...) (modern polars API)
• Removed all pandas imports and dependencies
• Updated exercise hints to idiomatic polars syntax

Terminology & Prose

• "melting" → "unpivoting"
• "NA"/"NaN" → "null" where appropriate
• Reframed stack/unstack section to focus on polars reshaping (no MultiIndex)
• Updated exercise prompts to reflect polars behavior

Files Modified

• data-tidy.ipynb

[brentomagic]

.

Approving after reconsidering — leftover melt wording is fine for now.

[brentomagic]

LGTM — approving the tidy-data Polars migration.