vtn/opencl: always lower to libclc fmod

The nir/spirv variant is simply not precise enough and almost everybody
lowers it anyway.

Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25837>

src/compiler/spirv/vtn_opencl.c

@@ -508,10 +508,6 @@ handle_special(struct vtn_builder *b, uint32_t opcode,
       return nir_cross3(nb, srcs[0], srcs[1]);
    case OpenCLstd_Fdim:
       return nir_fdim(nb, srcs[0], srcs[1]);
-   case OpenCLstd_Fmod:
-      if (nb->shader->options->lower_fmod)
-         break;
-      return nir_fmod(nb, srcs[0], srcs[1]);
    case OpenCLstd_Mad:
       return nir_fmad(nb, srcs[0], srcs[1], srcs[2]);
    case OpenCLstd_Maxmag: