Public Transit Demand Prediction During Highly Dynamic Conditions: A Meta-Analysis of State-of-the-Art Models and Open-Source Benchmarking Infrastructure