UNDO is complicated and it has the potential to destabilize the software (e.g. if we make a mistake, you hit undo and the software crashes). We have made the strategic decision to complete the functional feature set and focus on stability before implementing the undo infrastructure against the full implementation of the software.
But the good news is that we are getting very close to the end, and UNDO is one of the features that is on short list of items left to be implemented.
At this point your work-around is save or save as.