Feature/food review sentiment multilabel

[ertugrul-dmr]

sadedegel/prebuilt/README.md:

• Added how to use alternative prebuilt pipeline
• Added how to use evaluation
• Added holdout test set F-1 macro score

sadedegel/prebuilt/model/food_sentiment_multilabel.joblib:

• Dumped pretrained/optimized model weights for multilabel version

sadedegel/prebuilt/food_reviews_multilabel.py:

• Created functions according to CONTRIBUTING.md
• Implemented build, load and evaluate functions
• Added optional argument to check roc_auc score for predict_proba method
• Best model and vectorizer parameters are selected using optuna

tests/prebuilt/context.py:

• Added import line for the base model

tests/prebuilt/test_food_reviews_multilabel.py:

• Added assertions for testing purposes.

sadedegel/dataset/food_review/*

• Created a dataset imports by @dafajon based on the data I mentioned on Issue Food Delivery Reviews Classification Prebuilt Model #290
• Fixed minor dataset test case error

[dafajon]

@ertugrul-dmr Can you please update datasets/README.md with information on data source and the way you discretized the continious labels into class values.

[ertugrul-dmr]

Done @dafajon

[dafajon]

Thanks. Merging and closing.