RFC: Improve hypot performance

[Wilco Dijkstra]

Hi Paul,

> On Power10, this implementation is still has a large delta compared to the
> current implementation:

It's obvious something went wrong here since it is slower than Adhemerval's
version - did you look at the generated code?

It should use fma, and no division or function calls (if you see calls to fmin/fmax
you need to set FAST_FMINMAX to 0).

Cheers,
Wilco