Implement a proper database

[elifoster]

I think having a database where we store accessed user info, and tracking the generation time for those rows in order to only perform API requests once or twice a day per user could greatly increase our performance, and prevent API rate limit issues. @Strikingwolf what do you think?
User table:

PRIMARY KEY
Username    | Avatar | Current Streak | Longest Streak | Total Year | Average Day | Average Week | Average Month | Followers | Following | Public Repos | Private Repos | Forks   | Mirrors | Repos
String      | String | Integer        | Integer        | Integer    | Integer     | Integer      | Integer       | Integer   | Integer   | Integer      | Integer       | Integer | Integer | Pipe sep Strings

Repository table:

PRIMARY KEY
User/Repo   | Open Issues | Closed Issues | Open Pulls | Merged Pulls | Closed Pulls | Forks   | Stargazers | Watchers | Language Data
String      | Integer     | Integer       | Integer    | Integer      | Integer      | Integer | Integer    | Integer  | Hash as String

[FayeAlephNil]

Yes, definitely yes

It's fairly easy to implement as well using the Sequel gem

[elifoster]

I was reading about some database development stuff in pure Sinatra today, and was thinking of going with that. It won't be done until after #4 though, so it's still a while away.

[elifoster]

@Strikingwolf How do you suppose we'll save the actual language data? Right now in the main post I have it as a string that represents the main hash, but I hate that.

[elifoster]

The performance is officially so bad that Heroku will not load the site...

[elifoster]

@strikingwolf ping about the Lang data column