[ET-VK] Fix op benchmarks code generation (#6528)

## Context

As title. There was a small error in the operator benchmark code generation which impacted operators which return a collection of tensors (e.g. layer norm). The issue was that the generated code would store the output `ValueRefs` in variables that shadowed class variables of the benchmark.

Differential Revision: [D65071024](https://our.internmc.facebook.com/intern/diff/D65071024/)

ghstack-source-id: 250448840
Pull Request resolved: #6527

Co-authored-by: Stephen Jia <[email protected]>

Author: pytorchbot

backends/vulkan/test/op_tests/utils/gen_computegraph.py

@@ -448,7 +448,7 @@ def set_output(self, ref: ValueRefList, include_declarations: bool = True) -> st
         if isinstance(ref, list):
             ret_str = ""
             for r in ref[:-1]:
-                ret_str += self.set_output(r)
+                ret_str += self.set_output(r, include_declarations)
             return ret_str
         elif ref.src_cpp_type == TENSOR_VECTOR:
             assert ref.is_out