Bump version to 0.1.0 and add CHANGELOG

Version 0.1.0 introduces significant new functionality:
- Support for sklearn transformers and regressors
- New estimator parameter with backward-compatible clf parameter
- 100% test coverage
- Enhanced type detection and error handling

Add CHANGELOG.md to document all changes between versions.

Author: sepro

CHANGELOG.md

@@ -0,0 +1,55 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.1.0] - 2025-10-07
+
+### Added
+- Support for sklearn transformers (StandardScaler, PCA, MinMaxScaler, etc.)
+- Support for sklearn regressors (LinearRegression, Ridge, Lasso, etc.)
+- New `estimator` parameter for Model class (more intuitive than `clf`)
+- `transform()` method for transformer workflows
+- Type detection system to distinguish between classifiers, regressors, and transformers
+- Comprehensive error handling with helpful TypeError messages for incompatible operations
+- 100% test coverage across all source files
+- Transformer usage example in README
+
+### Changed
+- Model class now supports three types of estimators: classifiers, regressors, and transformers
+- `_get_estimator_type()` method returns estimator type as string
+- Summary now includes `estimator_type` field
+- Enhanced documentation in CLAUDE.md with usage examples for all estimator types
+
+### Fixed
+- Type safety: mixins now properly reject non-classifier estimators
+- Proper error messages when calling incompatible methods (e.g., predict on transformer)
+
+### Deprecated
+- None (full backward compatibility maintained)
+
+### Removed
+- None
+
+### Security
+- None
+
+### Notes
+- The `clf` parameter is still fully supported for backward compatibility
+- All existing pickled models from v0.0.2 load without any changes required
+- Evaluation for regressors is not yet implemented (raises NotImplementedError)
+
+## [0.0.2]
+
+### Added
+- Initial public release
+- Data class for train/test split management
+- Model class for classifier serialization
+- Confusion matrix visualization
+- ROC curve visualization
+- SHAP beeswarm plot support
+- Model evaluation metrics (accuracy, precision, recall, f1, roc_auc)
+- `purge()` method to remove data while keeping headers
+- `create_overview()` utility for model comparison

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name="ColdSnap",
-    version="0.0.2",
+    version="0.1.0",
     author="Sebastian Proost",
     author_email="sebastian.proost@gmail.com",
     description="Create snapshots of machine learning models and their training data",