Multisig actor update

[austinabell]

Summary of changes
Changes introduced in this pull request:

• Self explanatory. This one shouldn't be that bad to review

Reference issue to close (if applicable)

Closes #560

Other information and links

[dutterbutter]

This fn seems really weird to me, I know its 1:1 matching. It just seems weird that we would be returning an Ok exit code most of the time and wouldn't applied always be true?

[austinabell]

well the exit code is directly used in a serialized data structure, so I don't see the benefit in not returning the exit code (or how else would you pass that info back, because a result within a result doesn't seem like a good option. (If it's confusing, the ActorError in the return is different from this exit code, and if those three fields weren't directly used I would consider something else like an enum return)

And no applied is only true if the multisig threshold is met and that block of code gets executed

[dutterbutter]

I see, so if the multisig threshold is NOT met we still return an Ok exit code even though applied would be false?

[austinabell]

yeah, they leave as uninitialized, which is the 0 case (ExitCode::Ok) and return that (I think it's a bit weird that they don't make the exit code and return nullable, but I assume this may be more consistent and less likely to panic if things are changed)