PR for cleaning up wording around client-to-authenticator interface

If anyone feels the urge to read specs on a Friday afternoon (or on the weekend), it would be great to have opinions on this PR: https://github.com/w3c/webauthn/pull/154

It is an attempt to address some issues seen in an in-depth review and while there is little change to the IDL I'd like to make sure that it's compatible with the WG's mental models.