intel/cs: Rework the way thread local ID is handled

Previously, brw_nir_lower_intrinsics added the param and then emitted a
load_uniform intrinsic to load it directly.  This commit switches things
over to use a specific NIR intrinsic for the thread id.  The one thing I
don't like about this approach is that we have to copy thread_local_id
over to the new visitor in import_uniforms.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>

src/intel/compiler/brw_nir.h

@@ -95,8 +95,7 @@ void brw_nir_analyze_boolean_resolves(nir_shader *nir);
 nir_shader *brw_preprocess_nir(const struct brw_compiler *compiler,
                                nir_shader *nir);
 
-bool brw_nir_lower_cs_intrinsics(nir_shader *nir,
-                                 struct brw_cs_prog_data *prog_data);
+bool brw_nir_lower_cs_intrinsics(nir_shader *nir);
 void brw_nir_lower_vs_inputs(nir_shader *nir,
                              bool use_legacy_snorm_formula,
                              const uint8_t *vs_attrib_wa_flags);