aco/gfx11: perform FS input loads in WQM

aco/gfx11: perform FS input loads in WQM

fossil-db (gfx1100):
Totals from 48184 (35.68% of 135032) affected shaders:
MaxWaves: 1131876 -> 1131960 (+0.01%); split: +0.05%, -0.04%
Instrs: 36755466 -> 36782290 (+0.07%); split: -0.04%, +0.11%
CodeSize: 200812068 -> 200915348 (+0.05%); split: -0.04%, +0.09%
VGPRs: 2163980 -> 2163828 (-0.01%); split: -0.15%, +0.14%
Latency: 484174459 -> 484341018 (+0.03%); split: -0.06%, +0.09%
InvThroughput: 87941284 -> 87944874 (+0.00%); split: -0.04%, +0.04%
VClause: 652984 -> 653085 (+0.02%); split: -0.09%, +0.10%
SClause: 1510995 -> 1528832 (+1.18%); split: -0.40%, +1.58%
Copies: 1997689 -> 2001857 (+0.21%); split: -0.49%, +0.69%
Branches: 676629 -> 676584 (-0.01%); split: -0.02%, +0.01%
PreSGPRs: 2033070 -> 2036725 (+0.18%)
PreVGPRs: 1903922 -> 1903897 (-0.00%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 3730be9873d ("aco: mostly implement FS input loads on GFX11")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19370>