updated requirements, created docker file for HF deployment, modified lifespan to load model, stopped tracking joblib files

Add FastAPI backend wrapping ML inference pipeline