Develop crawler to collect challenges data

[tschaffter]

This Epic aims to explore the use of a crawler to collect challenges data from:

• a challenge platform that includes a large number of challenges (e.g. > 100)
• a challenge platform where challenge data are highly structured

A suitable target challenge platform could be CodaLab or Kaggle.

A good starting point is https://github.com/biothings/biothings.crawler