refactor: Move IO specifics out of Schema, Collection

[AndreasAlbertQC]

Motivation

The current implementation of parquet-based storage is strongly coupled to the implementation of the Schema and Collection classes. This makes it hard to reuse parts of the logic that would also be useful for other future storage backends. To prepare the implementation of such future storage backends, this PR refactors the current logic to introduce a clearer interface for storage backends.

Changes

• Introduced the notion of a StorageBackend, i.e. a piece of code encapsulating a way of storing data and metadata.
• Refactored the parquet storage logic to use the StorageBackend interface

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (d8916e8) to head (f68fbfc).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##              main      #109    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files           42        45     +3     
  Lines         2450      2573   +123     
==========================================
+ Hits          2450      2573   +123     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[AndreasAlbertQC]

ping @borchero

[borchero]

Just some cosmetic comments, thanks! 🚀

[AndreasAlbertQC]

Thanks @borchero! While refactoring into multiple files, I realized I had completely missed that FailureInfo also needs to be serialized. I implemented this analogously now.

[borchero]

Let's go :D