Need a "first-run" mode to populate initial entitites (from json files)

[missaugustina]

Using json files, dump to CSV and figure out the db tables from that. The github api has ratelimits that could complicate things. We should allow for doing api pulls at some point but for simplicity and just to get the data going, we'll just do manual pulls and save them to json files.

We might not need a special writer class for each of these since the data will be more consistent than it was for Events.

• [x] Manually create json files from curl requests on CLI
• [x] Document api calls used to create json files
• [x] Get all events
• [x] Get Labels
• [x] Get Milestones
• [x] Get Issues
• [x] Get Issue Comments
• [x] Get Pull Requests
• [x] Get Pull Request Comments
• [x] Get Commits