If a test has multiple scenarios, we currently throw an exception the first time we call toMatchSnapshot() with different data. But then we miss the opportunity to update the rest of the scenarios in the .snap.debug file.
It would be better to let the test finish, and then raise if necessary with all differences together.