I hope there is a test that checks all additions to the standard equivalence library are correct

I hope there is a test that checks all additions to the standard equivalence library are correct

yep, and it caught my off-by-global-phase error.

Are there any other equivalences in the equivalence library (or that we should add to it) that can be written more efficiently with a csx gate?

Pull Request Test Coverage Report for Build 1476434419

• 15 of 15 (100.0%) changed or added relevant lines in 1 file are covered.
• 3 unchanged lines in 1 file lost coverage.
• Overall coverage decreased (-0.0006%) to 82.541%

Totals
Change from base Build 1476251358:	-0.0006%
Covered Lines:	49849
Relevant Lines:	60393

💛 - Coveralls

I think we're ok without a changelog, unless we normally put one in for additional equivalences?

I think we are ok without one too.

Are there any other equivalences in the equivalence library (or that we should add to it) that can be written more efficiently with a csx gate?

This can be extended to multi-controlled toffolis, see pg 17 of barenco https://arxiv.org/pdf/quant-ph/9503016.pdf
Something to add as an enhancement. The XX decomposer is perhaps more useful for leveraging fractional gates, since it can be used to synthesize infinite operations. But it cannot act beyond 2-qubit gates, so I think both have their place.