Adds a non-optimal implementation of int.__divmod__ + tests

Change the logic to determine path to parse_ast.py, otherwise it breaks usage of pyston binary.

There's still vestiges of the libunwind experiment; I'm leaving those in even
though they're dead since I think we should move soon back to that approach.

Changed the indentation of pretty much the entire codebase.

It did some things that I don't like that seem not configurable,
but overall it seems like an improvement, and nice to have a
canonical format going forward.

Will commit the result in the next changeset.

Also, had to update to a newer version of LLVM to get some new clang-format options.
We don't necessarily have to use the same version of clang-format as the version of
llvm that we build it against, but if we can it keeps the build system simpler.

Release llvm_patches and libunwind_patches into the public domain so they can be incorporated into the upstream projects and their users

Uses module name when generating IR, for improved readability.

As I touched the module repr, I added some tests to it. Turned out that cpython normalizes the path of a module and uses the pyc file in the repr string. That part is ignored in this test.

Performance optimizations

reduces runtime of fannkuch.py by about 0.8 secs

Reduces fannkuch.py runtime on my machine from 25.5 secs to 16.5 secs.

Add support to list.count

Implement backtick (repr)

Rearrange BoxedClass members to reduce size slightly (8 Bytes)

Type specialice more runtime methods

Whoops. register specialiced int functions correctly

Type specialice buildin math support

add listAppendArrayInternal which appends an array of elements to a list