NetKAT parses signed 32/64bit integers; should be unsigned

[smolkaj]

The parser currently uses Int64.of_string and Int32.of_string to parse 64 bit and 32 bit integers, respectively. However, these functions parse signed integers in the range [-2^(x-1), 2^(x-1)-1] for x = 32,64. What we want are unsigned integers in the range [0,2^x-1]. The functions to_int and to_int32 in Frenetic_Fdd.ml suffer from a similar problem, see @calebvoss's comment in #517.

The deeper issue here is that we're using signed types int, int32, and int64 throughout the code base when really we should be using uint, uint32, and uint64. This would avoid bugs of this nature. Unfortunately, OCaml does not support these types natively, so we will need to resort to some library.

[smolkaj]

The stdint packages provides various unsigned integer types together with parsing functions etc., but they currently don't have support for s-expressions, see andrenth/ocaml-stdint#10.