Adding "IEEE Float" and "IEEE Double" to `ValueType` enumerator

[urc-ELIT]

Why doesn't ValueType have "IEEE Float" and "IEEE Double" options?

If I understand correctly how the Signal structure works, it is impossible to know if a signal is represented with floating-point representation. This is important for correctly decoding and incoding signals.

I can try to make a pull request for implementing this feature if there isn't any other way for doing this with the current code.