st/nir: Don't use GLSL IR's MOD_TO_FLOOR lowering when using NIR.

Both GLSL IR and NIR perform the same mod -> floor lowering for 32-bit
types.  But nir_lower_double_ops is slightly more defensive against
lowered drcp precision loss, and handles mod(x, x) = 0 directly.  This
works well...assuming nir_lower_double_ops actually gets an fmod op to
lower in the first place.

The previous patches enabled NIR-based lowering for the remaining
drivers, so we can stop using the GLSL IR lowering when using NIR.

Fixes KHR-GL45.gpu_shader_fp64.builtin.mod_dvec[234] on iris.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
10 jobs for !998 with fmod in 10 minutes and 16 seconds (queued for 50 seconds)
latest detached
Status Job ID Name Coverage
  Containers Build
passed #341675
debian

00:00:21

 
  Build+Test
passed #341677
meson-clang

00:08:23

passed #341681
meson-clover

00:05:25

passed #341680
meson-main

00:05:09

passed #341676
meson-swr-glvnd

00:05:47

passed #341682
meson-vulkan

00:01:54

passed #341684
scons-llvm

00:04:50

passed #341683
scons-nollvm

00:03:11

passed #341678
scons-swr

00:06:29

passed #341679
scons-win64

00:06:06