Use different hasher for collections

There are a few collection types that we use heavily which involve hashing: `HashMap`, `HashSet`, `IndexMap` and `IndexSet`. By default, they use a strong hashing algorithm which is slow but provides stronger security guarantees. You can, however, use different hashing algorithms if you want, so we could use a faster one. We don't need to worry about security in this context, so we also don't need to pay the performance penalty of a slower hasher.

More information here: https://nnethercote.github.io/perf-book/hashing.html#alternative-hashers

As always, we should benchmark carefully on a realistic workload.