purefunctor
formatter continued 01
I continued my routine of trying to parse purescript code and fixing what is missing or broken.
During my fixing I added more debug info to the tests and handled crashes more transparently so they'll be easier trouble shooted in the future - the code is a bit ugly but I think it does the jobb.
We also have trouble with infinet loops in the parser - I've found one of them and did something that might or might not be a fix. I check the number of tokens we've eaten and if it stays constant we're stuck and I simply break out of the loop - now this method is not that sophisticated but solved some of the problems I have.
I also noticed we don't check that we've parsed the entire file when we generate our tests, that's something that we should look into if we are striving for correctness.
Looks good so far, though I'd advise staying away from the combinators in favor of the new separated/repeat APIs like what #21 is currently doing.
