Comparing named vectors, ignoring order (again)

[wch]

This is the same as #473. That previous issue was indirectly considered resolved by the implementation expect_setequal, but that's not what I had in mind. expect_setequal tests that two sortable vectors have the same contents, but what I had in mind was that two named vectors (including lists, which aren't generally sortable) would be tested to have the same key-value pairs.

The idea is that a and b below would be considered to be equivalent:

a <- list(x="a", y=2)
b <- list(y=2, x="a")

But these would not:

a <- list(x="a", y=2)
c <- list(x="a", z=2)

It was closed before because @hadley thought that doing it well would require quite a bit of code, but I don't think that's necessarily true -- I had a basic implementation of it in the previous issue.

[hadley]

I'd definitely consider a PR if you can figure out a good name for it.

[kevinushey]

expect_same_members()? expect_mapequal()? (since we're sort of treating the list like a map in this case)

[wch]

I like expect_mapequal(). Although there is the odd case where a named vector has multiple entries with the same name. I'm not sure how we want to handle that.