run `specialization` benchmarks on v2

[alejandro-vaz]

we are trying to assess what performance benefits does specialization have

run cargo bench and cargo bench --features specialization, and open a PR with no changes and the table as description

if there are barely any / no performance improvements, it's rational to remove it, the cost in complexity we are paying is really high

[alejandro-vaz]

@TDecking you were the one who remade specialization a while ago, do you know if it is really worth keeping??

I remember my old benchmarks on #405 that never got merged and spec performance wasn't even that good, which makes me wonder whether it's necessary

other crates like arrayvec don't actually have a feature for specialization, so maybe we're just overengineering here