fix bugs, add new metrics and tasks

[v1xerunt]

1. fix mismatch between actual output and document descriptions in adacare, concare, agent and stagenet
2. add new metrics: outcome-specific mean absolute error, early prediction score from paper "A Comprehensive Benchmark for COVID-19 Predictive Modeling Using Electronic Health Records in Intensive Care"
3. add regression related metrics
4. add new task: outcome-specific length of stay prediction
5. change base_model to support new tasks