GH-18: feat(pipeline): Add serialization for pipeline scores

[Urfoex]

Adds Pydantic models and serialization logic for pipeline.scores. This includes support for both classification and regression scores, which are handled as a discriminated union.

• Defines ClassificationScore and RegressionScore Pydantic models.
• Implements serialize_scores to convert getML scores into the new models.
• Adds a custom WrongPipelineScoreTypeError exception for score type mismatches during serialization.
• Updates the PipelineInformation model and associated tests to include the new scores field.

Resolves #18

[Copilot]

Pull Request Overview

Adds serialization support for pipeline scores by implementing Pydantic models and serialization logic to handle both classification and regression score types as a discriminated union.

• Implements ClassificationScore and RegressionScore Pydantic models with type discrimination
• Adds serialize_scores function to convert getML scores into the new models with type validation
• Updates PipelineInformation model to include the new scores field with comprehensive test coverage

Reviewed Changes

Copilot reviewed 13 out of 13 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/getml_io/getml/scores.py	Defines Pydantic models for classification and regression scores with discriminated union
src/getml_io/serialize/pipeline.py	Implements score serialization logic with type validation and error handling
src/getml_io/serialize/exception.py	Adds custom exception for score type mismatches during serialization
src/getml_io/metadata/pipeline_information.py	Updates PipelineInformation model to include scores field
Test files	Comprehensive test coverage for score serialization and pipeline information updates

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[gemini-code-assist]

Code Review

This pull request introduces serialization for pipeline scores, a valuable addition. The implementation is well-structured, defining Pydantic models for different score types, adding serialization logic, and including comprehensive tests. My review focuses on a few areas for improvement: refining the Pydantic model for scores to be more idiomatic and flexible, optimizing the serialization functions for better performance and readability, and strengthening the test assertions to be more comprehensive.

[srnnkls]

Perfect.

[awaismirza92]

LGTM