intel/compiler: Be more conservative about subgroup sizes in GL

The rules for gl_SubgroupSize in Vulkan require that it be a constant
that can be queried through the API.  However, all GL requires is that
it's a uniform.  Instead of always claiming that the subgroup size in
the shader is 32 in GL like we have to do for Vulkan, claim 8 for
geometry stages, the maximum for fragment shaders, and the actual size
for compute.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>

src/intel/compiler/brw_vec4_tcs.cpp

@@ -397,7 +397,7 @@ brw_compile_tcs(const struct brw_compiler *compiler,
                             nir->info.outputs_written,
                             nir->info.patch_outputs_written);
 
-   brw_nir_apply_key(nir, compiler, &key->base, is_scalar);
+   brw_nir_apply_key(nir, compiler, &key->base, 8, is_scalar);
    brw_nir_lower_vue_inputs(nir, &input_vue_map);
    brw_nir_lower_tcs_outputs(nir, &vue_prog_data->vue_map,
                              key->tes_primitive_mode);