perf: provide items_per_second metric in bps_tree

This patch considers the number of iterations as the number of items
proceeded by the corresponding benchmark, so it may be used for the
`items_per_second` counter.

NO_DOC=perf test
NO_CHANGELOG=perf test
NO_TEST=perf test

perf/bps_tree.cc

@@ -120,6 +120,7 @@
 	type##_##func##_size_##size(benchmark::State &state) \
 	{ \
 		test_##func<type>(state, size); \
+		state.SetItemsProcessed(size); \
 	} \
 	BENCHMARK(type##_##func##_size_##size)
 
@@ -128,6 +129,7 @@
 	type##_##func##_height_##height(benchmark::State &state) \
 	{ \
 		test_##func<type>(state, type::height_##height##_max_size); \
+		state.SetItemsProcessed(type::height_##height##_max_size); \
 	} \
 	BENCHMARK(type##_##func##_height_##height)