iris: Ask st to vectorize our IO.

(Technically this is common code, but it doesn't affect i965 or anv.)

Improves performance of GFXBench5's gl_tess benchmark on Skylake GT4e
by 12.1418% +/- 0.159919% (n=3) by eliminating all spilling in the GS.
10 jobs for !644 with st-vectorize in 10 minutes and 37 seconds (queued for 1 second)
detached
Status Job ID Name Coverage
  Containers Build
passed #311456
debian

00:00:23

 
  Build+Test
passed #311458
meson-clang

00:10:14

passed #311462
meson-clover

00:07:09

passed #311461
meson-main

00:05:27

passed #311457
meson-swr-glvnd

00:07:53

passed #311463
meson-vulkan

00:02:24

passed #311465
scons-llvm

00:05:16

passed #311464
scons-nollvm

00:03:24

passed #311459
scons-swr

00:05:26

passed #311460
scons-win64

00:08:23