Failing inheritance fixes

[JakeGinnivan]

Builds on @rcknight's pull request at #429. This fixed the perf issues when inheriting branch config and also fixes the failing test in #429

@asbjornu @GeertvanHorrik fyi and code review would be grand

@nulltoken could you have a look at the libgit2 stuff?

[GeertvanHorrik]

Why using regex for this one? Maybe it's even better to make this a setting (I see some people using "dev" instead on the inet).

[asbjornu]

A setting for this is already available (as per #292), no?

[GeertvanHorrik]

Then we should use it :)

[JakeGinnivan]

I think this can be removed. The new logic should allow not having this fallback. Let me look into it

[JakeGinnivan]

This can't be removed at the moment. If the code gets to the point where it cannot figure out which branch to inherit from then this code needs to be able to pick between master and develop.

An example:

* [master, develop]
|
* [feature/foo]

In this scenario there is no way to know if GitVersion should inheirit the config from master or develop, this code makes sure that develop gets selected. It isn't great but I can't think of a better solution?

[GeertvanHorrik]

I think it looks good, but I am not 100 % into GitVersion (yet), so might have overseen something.