Consider adding benchmarks that surface the performance difference between eager_global_ordinals and murmur3

[eddieturizo]

I'm not sure how often this comes up but we might want to consider running benchmarks between eager_global_ordinals and the murmur3 plugin as it relates to the performance differences of each when used for terms aggregations against high cardinality fields.

[eddieturizo]

Given the use case disparity (terms aggs vs cardinality aggs) between global ordinals and murmur3, it may not make sense to run benchmarks comparing the two. It may make more sense to generate independent benchmarks for each.