Fixing rebase rather than avoiding it

[Stefan Monnier]

> I certainly wouldn't expect anyone to be allowed to submit a change to
> public branch that changed it's published history but I would expect

I definitely would.  Of course only in particular cases:
- modifying commit-log messages.
- adding parents to rev-ids.
- suppress some old (non-tip) revisions whose content is unacceptable.
  This will require special support to make sure that people can still
  "bzr pull" from a branch that happens to have one of those suppressed
  revisions as its tip.  E.g. the rev-id of those suppressed revisions
  needs to be kept in the history.


        Stefan