Add Clinical Firewall HARD-LOCK + WhatIfScenarioCard frontend

feat: add diabetes model artifacts, training pipelines, and feature engineering

feat: add diabetes model artifacts, training pipelines, and feature engineering

chore: add model diagnostic artifacts and evaluation plots

fix: add missing engineer_medical method to HeartDiseaseFeatureEngineer + update preprocessors

fix: StringDtype bug in engineer_medical_features + retrain model for 16-feature consistency

Delete models/heart_disease/final_model.pkl with huggingface_hub

Delete models/final_model.pkl with huggingface_hub

Delete models/omni_diag_xgb_optimized.pkl with huggingface_hub

fix: resolve model path and finalize project restructuring

chore: remove binary files rejected by Hugging Face Spaces

chore(models): add isolated model artifacts for heart_disease module

refactor(models): migrate model training scripts to config-driven paths

feat(models): encapsulate feature engineering logic into reusable functions

feat: add SHAP explanation and patient report generation in live demo

feat: add optimized XGBoost and TabNet model weights for improved performance

feat: enhance ensemble prediction script with improved data handling and model integration

feat: implement TabNet model training with data preprocessing and evaluation

fix: explicitly define target column as 'HeartDisease' and add validation check

fix: correct parameter name in train_test_split function for data splitting

feat: update XGBoost training script with optimized parameters and data handling improvements

feat: add experiments log for A/B testing results of feature engineering approaches

feat: add ensemble prediction script integrating TabNet and XGBoost models with soft voting

feat: update XGBoost optimization to evaluate new engineered features using both data_clinical and data_heuristic

feat: add XGBoost training script with optimized parameters and model saving

feat: implement advanced feature engineering for heuristic and clinical risk scoring

feat: add model weights for TabNet and XGBoost training

feat: add XGBoost optimization script with Optuna for hyperparameter tuning

feat: implement XGBoost training pipeline with MLflow logging and model saving

feat: implement TabNet training pipeline with MLflow integration and model saving

Add label encoders and standard scaler as binary files for preprocessing

Refactor project structure to standard Data Science architecture

Force retrain model to overwrite old version

Merge remote skeleton into local project

Initial commit - Heart Disease 1 Project