I think this is a good idea. In my opinion it would be authors’ responsibility not to abuse the new ability: the merge directive should be added only once the PR has reached an agreement. That is a subjective judgment, but it would be hard to codify it.
One question—what should happen if (1) PR is approved and looks non-controvesial, (2) author adds a merge directive, and then (3) some other reviewer requests changes? Should we expect the merge directive to be canceled? I think we should consider what is actually possible to implement in automation, maybe there is no way to handle this scenario.