Test IterableSubject methods on objects that don't implement hashCode()

[cpovirk]

Such objects are at least a little evil, but they exist in one or two Google-internal projects. If we currently support this, then we should continue to do so, or at least we should have a test that lets us know that we're breaking it so that we don't do so by mistake.

[cpovirk]

By "don't implement hashCode()", I was thinking more of classes that implement it by throwing UnsupportedOperationException. But I just realized that the particular case I was worse, overriding equals() but not hashCode(). But I suppose that a test for the UnsupportedOperationException will test that, too, for better or for worse.

[cpovirk]

We just got a report internally that failing tests still call hashCode here: https://github.com/google/truth/blob/f149531ce3b6e69f68344ac016b4980f9a10744c/core/src/main/java/com/google/common/truth/SubjectUtils.java#L267

I wonder if we could get away with using an ArrayListMultimap (which I don't think would use hashCode) instead of a HashMultimap?

[Update: I did that.]