When copying a MatrixBasedOperator, the functors are also copied, in
particular the functor which is appended in the constructors. However,
the copied functor still used the assembled_ member of the
copied-from MatrixBasedOperator, and not the one of the new
MatrixBasedOperator.

As long as we are using these everywhere in our own code, we should not
warn about it. Otherwise, we have a lot of these warnings which hide
relevant other warnings.