math: optimize ceil/floor functions on amd64

math: optimize ceil/floor functions on amd64

Use SSE 4.1 rounding instruction to perform rounding
Results (haswell):

name      old time/op  new time/op  delta
Floor-48  2.71ns ± 0%  1.87ns ± 1%  -31.17%  (p=0.000 n=16+19)
Ceil-48   3.09ns ± 3%  2.16ns ± 0%  -30.16%  (p=0.000 n=19+12)

Change-Id: If63715879eed6530b1eb4fc96132d827f8f43909
Reviewed-on: https://go-review.googlesource.com/14561
Reviewed-by: Klaus Post <klauspost@gmail.com>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>