[PROFILES] Add contributors and watchers

[jeremyharris]

It would be nice to see at least the top contributors and watchers of a project. It might help to give people a sense of belonging, too. Something similar to: http://www.tus.io/about.html

[josegonzalez]

For the following repo:

• https://api.github.com/repos/josegonzalez/page_route

We can have the following data:

• Stargazers: https://api.github.com/repos/josegonzalez/page_route/stargazers
• Subscribers: https://api.github.com/repos/josegonzalez/page_route/subscribers
• Contributors: https://api.github.com/repos/josegonzalez/page_route/contributors
• Collaborators: https://api.github.com/repos/josegonzalez/page_route/collaborators

I think perhaps collaborators/contributors can be combined, and the same applies to stargazers/subscribers.

So that we don't have to constantly re-crawl these, we can have a pass-thru cache. It would work as follows:

1. Someone goes to http://plugins.cakephp.org/p/1231-HipHop
2. JS fires to retrieve http://plugins.cakephp.org/packages/1231/contributors and http://plugins.cakephp.org/packages/1231/watchers
3. Cakepackages either:
   • Retrieves the cached content from the db
   • Inline retrieves the data and responds. This is slow, so maybe we don't always do this.
4. If a db retrieval fires, then we also fire a job that will update the data if it hasn't been updated within X days

This might require reworking of the package page. I'll try and get to a mockup tomorrow and you can let me know if you'd change anything.

[jeremyharris]

Sounds good!

I was going to suggest using https://github.com/jeremyharris/cacher to help with the caching, but it looks like your Github datasource already handles it. If you want to extract that logic from the datasource, or cache other query results, it may help ease some of your dev time. Just a thought.

Thanks for the hard work @josegonzalez.