Alternative definition of `no_new_unsolved_evars` tactic

Following a suggestion by @jung in !429 (diffs, comment 86725)

@jung Could you run CI on all reverse dependencies to see if this indeed does not break anything?

Todo:

• Also improve naming to make sure this tactic either solves or fails.

mentioned in merge request !429 (merged)

Could you run CI on all reverse dependencies to see if this indeed does not break anything?

I can, once Iron CI is fixed. ;)

EDIT: Ah, it has been fixed. :) CI is running.

I didn't get an error email so looks like the reverse dependencies all keep working.

merged

mentioned in commit 469711e5

Should this go into the changelog, just in case?

I don't think so, we expected the behavior to be equivalent, and the CI run seems to confirm that.