What about ferromagnetic fluid? You can still spin it if you are aiming to push it 'centrifugally' into the features of the vacuum boundary, but you shouldn't lose a lot during the detachment.
(Plus it looks much more sci-fi.)
(Plus it looks much more sci-fi.)
I wonder if some sort of jelly would work to do the same thing, like a snail. Or a dynamic cup surface that conforms to the wall better.