content API update and some other features

[whoisterencelee]

Instead of using read/write, the content API should be used unless the data is more than 1MB, because the number of http requests get reduced from 3 to just 1, and allows multi-user commits which tracks the document SHA (prevents update with a old SHA).

A few changes to allow 4 things:

1. fix for issue Fix the contents function on the repository API. #101
   getting Content(branch, path, cb, sync, 'json') will get a json with file obj sha, which is closer to what http://developer.github.com/v3/repos/contents/#get-contents is describing, with option 'json' will give the same behaviour as before, getting the file raw content

2. added a repo.req function to allow access to request, which is useful if anyone wants to make custom API calls

3. allow request status==0 which is returned from XHR request made from a locally loaded page

4. getCommits should use pagination

[aendra-rininsland]

@whoisterencelee Would you mind solving the merge conflicts and rebasing if you still would like this added? It seems like a worthwhile PR and I'd like to add it to 0.10.8 if possible. Apologies for the delay!

[aendra-rininsland]

Actually, reviewing this a bit more, it's effectively duplicating existing PRs. Closing as duplicate.