feat: Introduce pydantic conversion for schemas

[Claude]

Motivation

Closes #323.

Changes

• Add to_pydantic_model on dy.Schema

Side-note: GitHub-orchestrated Claude was really unhelpful 🫠

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (f5de4fc) to head (6a06e13).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##              main      #324    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files           56        56            
  Lines         3233      3399   +166     
==========================================
+ Hits          3233      3399   +166     

☔ 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.

[Oliver Borchert (borchero)]

@claude[agent] implement all review comments

[Oliver Borchert (borchero)]

@claude[agent] implement the review comment

[Copilot]

Pull request overview

Adds first-class Pydantic v2 support to Dataframely schemas/columns to enable generating structured-output models (e.g., for LLM constrained decoding) directly from dy.Schema.

Changes:

• Add Schema.to_pydantic_model() to generate a pydantic.BaseModel reflecting schema columns and structured constraints.
• Introduce Column.pydantic_field() plus per-column _python_type / _pydantic_field_kwargs() implementations to translate constraints into Pydantic annotations.
• Add an optionals-only test suite covering pydantic field generation and basic schema-to-model conversion.

Reviewed changes

Copilot reviewed 19 out of 19 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tests/schema/test_pydantic_model.py	Tests for Schema.to_pydantic_model() behavior (basic fields, validation, warning on schema rules).
tests/columns/test_pydantic.py	Tests for Column.pydantic_field() across many column types and constraints/warnings.
dataframely/schema.py	Implements Schema.to_pydantic_model() using pydantic.create_model.
dataframely/columns/_base.py	Adds Column.pydantic_field(), _python_type abstract property, and _pydantic_field_kwargs() hook.
dataframely/columns/_mixins.py	Adds _pydantic_field_kwargs() to OrdinalMixin to translate min/max constraints.
dataframely/columns/any.py	Defines _python_type for Any to support pydantic type generation.
dataframely/columns/array.py	Adds _python_type and pydantic length constraints for arrays (incl. multidim warning).
dataframely/columns/binary.py	Defines _python_type for binary columns.
dataframely/columns/bool.py	Defines _python_type for bool columns.
dataframely/columns/categorical.py	Defines _python_type for categorical columns.
dataframely/columns/datetime.py	Defines _python_type and warnings for non-translatable datetime/date/time/duration constraints.
dataframely/columns/decimal.py	Defines _python_type and pydantic decimal-place constraint.
dataframely/columns/enum.py	Defines _python_type using Literal[...] for category validation.
dataframely/columns/float.py	Defines _python_type and pydantic constraints for bounds/inf/nan handling (incl. warning).
dataframely/columns/integer.py	Defines _python_type and pydantic constraints for integer bounds and is_in.
dataframely/columns/list.py	Defines _python_type and pydantic min/max length constraints for lists.
dataframely/columns/object.py	Defines _python_type for object columns.
dataframely/columns/string.py	Defines _python_type and pydantic length/pattern constraints for strings.
dataframely/columns/struct.py	Defines _python_type as an inner pydantic model for struct validation.

[Andreas Albert (AndreasAlbertQC)]

Should this be configurable? Our users may or may not use the ...Schema naming convention

[Oliver Borchert (borchero)]

Do you want to allow the user to pass a name? If their schema name does not end in Schema, this doesn't have any effect.

[Andreas Albert (AndreasAlbertQC)]

This will create a ton of warnings if you have a few custom checks. Do we need to warn on execution as opposed to just being really clear about this in the docuemntation of to_pydantic_model?

[Oliver Borchert (borchero)]

We could have all these warnings have the same prefix so that they're easy to ignore? 🤔 not sure if you can shoot yourself in the foot too easily without the warnings