Support export of new ML models

[StrikerRUS]

This is meta-issue to consolidate all requests for supporting export of new machine learning models in one place.

• [scikit-learn] Gradient Boosting (Add support for Gradient Boosting from scikit-learn #232)
• [scikit-learn] Hist Gradient Boosting (Add support for Hist Gradient Boosting from scikit-learn #381)
• [scikit-learn] Multi-target Regression (Add support for multi-target regression #280)
• [scikit-learn] Multi-label Classification (for DecisionTreeClassifier) (Add support for multi-label classification of DecisionTreeClassifier(sklearn) #479)
• [scikit-learn] CalibratedClassifierCV (Add support for CalibratedClassifierCV from scikit-learn #462)
• [scikit-learn] Isolation Forest (Add support for Isolation Forest #330)
• [scikit-learn] Naive Bayes (Add support for Naive Bayes #350)
• [scikit-learn] Stacking (Add support for stacking #368)
• [scikit-learn] KNeighbors (Add support for KNeighbors from scikit-learn #415)
• [scikit-learn] MLP (Add support for MLP from scikit-learn #431)
• [catboost] CatBoost (Add support for CatBoost #347)
• [xgboost] New XGBoost JSON schema (Add support for new XGBoost JSON schema #396)
• [ebm] Explainable Boosting Machine (Add support for Explainable Boosting Machine #429)
• [statsmodels] More different models (Add support for more models from statsmodels #384)