feat: add presentationidx metric

feat: implement BestChunkMatch metric

style: indent prompts

feat: log n questions

feat: configure logging all params, move preprocessing to be a feature of a storage

feat: fix async eval to be really async

feat: configure cli and implement saving eval results to gsheets

feat: configure benchmarking and add scorer pick to cli

feat: add async support in evaluation. partially add uploading results to gsheets

feat: implement retrieval with llm reranking

feat: add logging to evaluation

refactor: add oop to make metrics consistent

feat: setup mlflow evaluation

feat: add mlflow evaluation

feat: setup metrics. add llm relevance

feat: implement model evaluation