Return an accurate floating-point sum of values in the iterable seq. Assumes IEEE-754 floating-point arithmetic.
math/fsum
fsum