Unique Index format is going to change as part of the work planned for this epic. It would be useful to expand Evergreen microbenchmark tests to cover unique index better. Currently there is query benchmark that uses unique indexes. We should add cases to insert and update benchmark that similarly use unique indexes.