The 'oversleep' case

While Princeton WN is reluctant to have 'sleep' as hypernym/troponym of 'oversleep' (just retaining the entailment relation), this PR has 'sleep' as the hypernym of 'oversleep'. IMO this is questionable.

In the WN glossary of terms wngloss(7WN), a troponym is defined as "a verb expressing a specific manner elaboration of another verb. X is a troponym of Y if to X is to Y in some manner."

When applying to the 'oversleep'/'sleep' relation, it does not really make sense to say that to oversleep is to sleep in some manner.

However it is to 'fail' in some manner: fail to wake up as intended or in time... The same fail-to feature applies to some verbs built with 'over-' prefix, most verbs with 'mis-' prefix like 'misfire'.

Some entail the base verb (oversleep, overreact), some do not (misfire, misunderstand).

To support this view, note that negation seems to bear on the fail-to feature:
from "he didn't oversleep" one cannot jump to "he didn't sleep",

I am rejecting this (I know it is my own work).

I don't think on reflection that the work is of the quality to be accepted so I will revise this before resubmitting.