Clang-based C parser

[h0nzZik]

Also includes #577.

[dwightguth]

This seems okay to me. I didn't really look in detail at the specifics of the parser or at the semantics code though, so someone else should still review the semantic changes. (@mickyabir or @ehildenb probably).

I assume you have gone over the list of failing tests and ensured that none of them are due to bugs in your code?

[h0nzZik]

Oh sorry, I did not notice the comment and merged it. @mickyabir @ehildenb Guys, can you look at the semantics changes?
@dwightguth ./offsetof-1.c was failing in Clang, but after Chris fixed the test, it is failing in this PR because it was not implemented in the Clang-based parser. I have already implemented it locally, but it is now failing due to the problem with OffsetOfNode::getArrayExprIndex that I mentioned on slack. I hope to enable that test in the next PR.