feat(Smt): Reject duplicate keys in inputs

[iamrecursion]

Describe your changes

This change makes it so all SMT operations that operate on batches of key-value pairs will reject any input where the same key occurs more than once. This ensures that all of the various SMT implementations are coherent on the results they provide as there are no longer order-dependent computations.

This commit introduces no measurable performance change when it comes to the concurrent and sequential compute_mutations versions, and LargeSmt does not suffer any performance regressions either.

I've left LargeSmtForest out of this PR for the simple reason that it already has its own way of handling this. Its input batches SmtUpdateBatch and ForestUpdateBatch always deduplicate and retain the last operation for any key. Please let me know if you'd like to see the forest return an error instead. It's a fairly trivial change.

Performance-wise, best I can tell this causes no regressions. I am having some strange issue with the benchmarks today where some operations will regress and others will stay the same, and then when I re-run others will regress and the prior regressions will go back to normal. If somebody reviewing this could run the following benchmarks before and after this commit to confirm my findings here I would be most appreciative.

• smt.rs: smt_compute_mutations smt_apply_mutations smt_batch_updates smt_insert
• large_smt.rs: large_smt_compute_mutations large_smt_apply_mutations large_smt_insert_batch large_smt_insert_batch_to_empty_tree large_smt_insert_batch_to_populated_tree memory_smt_compute_mutations memory_smt_apply_mutations memory_smt_insert_batch

Closes #867.

Checklist before requesting a review

• Repo forked and branch created from next according to naming convention.
• Commit messages and codestyle follow conventions.
• Relevant issues are linked in the PR description.
• Tests added for new functionality.
• Documentation/comments updated according to changes.

[huitseeker]

I think the duplicate-key rejection can be bypassed in batch paths, see inline.

[iamrecursion]

Good catch on that. I had it in my head that I was sorting directly by key (which implents the behaviour that I want), rather than leaf index. That is now what I am doing.