Revert "intel/fs: only avoid SIMD32 if strictly inferior in throughput"

This reverts commit 6b494745be.

The logic is not entirely correct: the comparison is between two
static-analysis estimates of a dynamic system with variables that aren't
captured by the shader source, so using ">" will always have greater potential
to cause regressions whenever the performance difference between the two builds
is something not captured by the static model, no matter how much the model is
improved.

Reference: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9262
(cherry picked from commit d142c845d0)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25377>

src/intel/compiler/brw_fs.cpp

@@ -7678,7 +7678,7 @@ brw_compile_fs(const struct brw_compiler *compiler,
       } else {
          const performance &perf = v32->performance_analysis.require();
 
-         if (!INTEL_DEBUG(DEBUG_DO32) && throughput > perf.throughput) {
+         if (!INTEL_DEBUG(DEBUG_DO32) && throughput >= perf.throughput) {
             brw_shader_perf_log(compiler, params->log_data,
                                 "SIMD32 shader inefficient\n");
          } else {