Add Serialization for khash map

[dnbaker]

I've tested this on a map with 4096 keys, and kh_exist, kh_key, and kh_val all agreed between original and loaded maps. This only works, however, with maps, not sets.

[dnbaker]

I've updated this to work for sets as well as maps.

[justinmk]

It looks like this is writing bits to a file in a specific order, not "stringifying" the khash contents. (In case anyone else was wondering.)

Also, why write to a filepath instead of a caller-supplied buffer?

[justinmk]

This duplicates a lot of code and it's not clear how it's related to the serialization.

kputl could call kputl_.

[dnbaker]

This is unrelated to serialization. I added this so that I'd have the option of integer formatting routines which did not null-terminate for cases where I knew I would be appending to the string further. I can separate that out from this pull request later if requested. Compare kputw and kputw_ which were already present, for example. I'm not quite sure I understand why only some types were provided separate functions.

[dnbaker]

That's a reasonable point. I think having separate write (for writing to a buffer) and serialize methods (for writing to a file) where the latter calls the former is probably best. This also does not work for string sets or maps, but external storage complicates things.

[dnbaker]

I've updated this. kh_read_##name and kh_write_##name both work on FILE *s, while kh_serialize_##name and kh_deserialize_##name operate on paths.