Test expected failing or unhandled cases

[ijpulidos]

This set of changes is aimed at writing test cases that are expected to fail in the protein mutation protocol, either because of limitations of the methodology or because these are simply cases that we don't support to date.

Solves #80

[ijpulidos]

As usual, just bear in mind that these tests will fail as of today. We will be implementing the protocol to comply to these tests eventually. Some tweaks might be needed when that happens, but the idea is that the overall shape of these tests stay the same when we implement the protein mutation protocol.

[IAlibay]

is there a reason generally why we want to avoid

ring growing

Handling ring breaking is something that the current hybrid topology cannot handle properly - it's a fundamental issue in the way in which we deal with bond scaling. There are some ways around this, but they would require a rather large rewrite of the hybrid topology approach and generally isn't something we want to invest a lot of our time on.

double charge changes

This is related to https://github.com/OpenFreeEnergy/openfe/issues/660

The main answer here is "the tooling doesn't allow it now, we could do it, but it's not the best use of our resources".

[ijpulidos]

I renamed some of the custom errors and rewrote some of the strings, I think we could probably reiterate on these if needed in the future. I'll go ahead and merge this.