Design protocol classes and update architecture for model comparison analysis

[Copilot]

This PR introduces model comparison analysis capabilities to tab-right by implementing two new protocol classes that follow the existing architectural patterns.

Changes Made

New Protocol Classes

1. Prediction Calculation Protocol (model_comparison_protocols.py)

• PredictionCalculationP: Protocol for calculating pointwise errors between multiple prediction datasets and true labels
• Initialized with DataFrame and label column name
• __call__ method accepts list of prediction Series and optional error function
• Calculates pointwise error for each prediction against the label using specified or default error function

2. Plotting Protocol (model_comparison_plot_protocols.py)

• ModelComparisonPlotP: Protocol for visualizing comparison results between multiple predictions and labels
• Provides interfaces for error distribution plots, pairwise comparisons, and performance summaries
• Follows existing plotting protocol patterns with Figure type unions (matplotlib/plotly)

Architecture Integration

• Protocols follow the established pattern used by segmentation and drift analysis modules
• Uses @runtime_checkable and @dataclass decorators for consistency
• Maintains the same documentation and type hinting standards
• Added to API documentation for discoverability

Testing

• Created comprehensive test suite with 9 new tests following existing patterns
• Added protocol compliance tests with dummy implementations
• All 145 tests pass, ensuring no regression in existing functionality
• Tests verify protocol interfaces, attributes, and method signatures

Example Usage

from tab_right.base_architecture.model_comparison_protocols import PredictionCalculationP
from tab_right.base_architecture.model_comparison_plot_protocols import ModelComparisonPlotP

# Protocol defines interface for comparing multiple model predictions
calc: PredictionCalculationP = MyModelComparison(df, "target_column")
results = calc([pred_series_1, pred_series_2], error_func=custom_metric)

# Protocol defines interface for visualizing comparisons  
plotter: ModelComparisonPlotP = MyComparisonPlotter(calc)
fig = plotter.plot_error_distribution([pred_series_1, pred_series_2])

The architecture now supports extensible model comparison analysis alongside existing segmentation and drift detection capabilities, providing a foundation for future analysis modules.

Fixes #109.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 23 lines in your changes missing coverage. Please review.

Project coverage is 92.88%. Comparing base (4c432a1) to head (09cd69c).
Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
...se_architecture/model_comparison_plot_protocols.py	0.00%	14 Missing ⚠️
...ht/base_architecture/model_comparison_protocols.py	0.00%	9 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #110      +/-   ##
==========================================
- Coverage   95.78%   92.88%   -2.91%     
==========================================
  Files          19       21       +2     
  Lines         736      759      +23     
==========================================
  Hits          705      705              
- Misses         31       54      +23     

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

[EH-MLS]

@copilot

[EH-MLS]

Suggested change

...

[EH-MLS]

Suggested change

...

[EH-MLS]

Suggested change

...

[EH-MLS]

remove this file for now

[Copilot]

Removed test file as requested in 3b8c1a4

[EH-MLS]

remove this file for now

[Copilot]

Removed test file as requested in 3b8c1a4

[EH-MLS]

@copilot
ci faling

[Copilot]

@copilot
ci faling

Fixed CI failures by resolving linting and type checking issues. All pre-commit hooks, tests (136/136), MyPy type checking, and documentation builds now pass. Commit: 09cd69c