Handle to float64 to float32 overflow

[ccoVeille]

#14 allowed float32/64 to all integer types.

As far as I know the only other overflow I'm not supporting is float64 to float32.

So ToFloat32 could be added.

Everything can be stored into a float64, so no overflow, but I can add ToFloat64 that will do nothing else than casting to fooat64

So, ToFloat64 could be added

Originally posted by @ccoVeille in #14 (comment)

This issue is related to #7