cmd/compile/internal/gc: speed-up small array comparison

cmd/compile/internal/gc: speed-up small array comparison

Currently we inline array comparisons for arrays with at most 4 elements.
Compare arrays with small size, but more than 4 elements (e. g. [16]byte)
with larger compares. This provides very slightly smaller binaries,
and results in faster code.

ArrayEqual-6  7.41ns ± 0%  3.17ns ± 0%  -57.15%  (p=0.000 n=10+10)

For go tool:
global text (code) = -559 bytes (-0.014566%)

This also helps mapaccess1_faststr, and maps in general:

MapDelete/Str/1-6               195ns ± 1%     186ns ± 2%   -4.47%  (p=0.000 n=10+10)
MapDelete/Str/2-6               211ns ± 1%     177ns ± 1%  -16.01%  (p=0.000 n=10+10)
MapDelete/Str/4-6               225ns ± 1%     183ns ± 1%  -18.49%  (p=0.000 n=8+10)
MapStringKeysEight_16-6        31.3ns ± 0%    28.6ns ± 0%   -8.63%  (p=0.000 n=6+9)
MapStringKeysEight_32-6        29.2ns ± 0%    27.6ns ± 0%   -5.45%  (p=0.000 n=10+10)
MapStringKeysEight_64-6        29.1ns ± 1%    27.5ns ± 0%   -5.46%  (p=0.000 n=10+10)
MapStringKeysEight_1M-6        29.1ns ± 1%    27.6ns ± 0%   -5.49%  (p=0.000 n=10+10)

Change-Id: I9ec98e41b233031e0e96c4e13d86a324f628ed4a
Reviewed-on: https://go-review.googlesource.com/40771
Run-TryBot: Ilya Tocar <ilya.tocar@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>