The library can now parse integers, strings and lists in "Bencoding" format (check the RFC).
I added a test case as well, which reveals a segfault whenever an incorrect char is parsed (this should be fixed obviously). I also don't free a single bit of memory yet (I know, I know...), so we need a proper way to free a whole bencoding data struct.

Something hard as well: NAMING. functions are "bencoding_*", variables/types "be*"... We need consistency, but I don't feel like using the full format name for variables. It will become unreadable.
As the next sections are THP and PWP, I was considering "ben" as a prefix for everything. I'm just wondering if that makes sense...