Experimental (unstable) “fast-math” wrappers for f32, f64

These wrappers enable the [“fast-math”][1] flags for the operations
where there are intrinsics for this (add, sub, mul, div, rem).
The wrappers exist so that we have a quick & easy way to experiment
with fast math flags and further that feature in Rust.

Note that as of this writing, the Rust instrinsics use the “fast” flag
documented in the langref; this enables all the float flags.

[1]: http://llvm.org/docs/LangRef.html#fast-math-flags

# Rust Version

This crate is nightly only and experimental. Breaking changes can occur at
any time, if changes in Rust require it.