Optimize scikit-learn pipeline and add --scikit_model_name flag

[vile319]

Optimizes the scikit-learn pipeline to handle large datasets and adds the ability to skip LazyPredict when a model is already known.

Changes
main.py

• When --scikit_model_name is specified, skips LazyPredict and goes directly to hyperparameter tuning + training

lazy_predict.py

• Precompute preprocessing once instead of refitting StandardScaler/Imputer per model
• Added n_jobs=-1 to parallelizable models (RandomForest, etc.)
• Removed slow models from LazyPredict: SVC, NuSVC, AdaBoost, KNeighbors, DecisionTree, LDA/QDA, etc.
• Added XGBoost/LightGBM to model dictionaries correctly

scikit_classes.py

• Fixed --scikit_model_name CLI arg mapping to model_name
• Added hyperparameter tuning when using --scikit_model_name directly
• Added verbose logging to RandomizedSearchCV

New Usage
Full pipeline (LazyPredict → best model → hyperparameter tuning)
python main.py --model_names ESMC-600 --data_names gold-ppi --embedding_pooling_types mean var --use_scikit --n_jobs -1

Skip LazyPredict, go straight to XGBoost tuning
python main.py --model_names ESMC-600 --data_names gold-ppi --embedding_pooling_types mean var --use_scikit --n_jobs -1 --scikit_model_name XGBClassifier --scikit_n_iter 10