* Split + buff succubus
When all succubi die, all entranced people now die along with them. This
should prevent an entranced person from ratting out the succubus early
on so that they go back to their team, as they lose now even if succubus
dies. One exception is if EVERY succubus idles out, then everyone that
is entranced is freed of entrancement, as it isn't their fault that they
didn't protect their friends in that case.
Dullahans now have succubi entirely removed from their list as the
likelihood they get unentranced is low, and it's easier to implement
this way.
Ensure that entranced people can vote along with ANY succubus, even if
that vote isn't the one that succeeded. Before there were cases where
they could vote along with succubus but still end up dying (particularly
in respect to a vote passing when a succubus abstained).
Clear up some message wording with regards to succubi.
Cleaned up chk_win_conditions and eliminated chk_traitor, so they make
much more sense now.
Also fixed minor issues, such as end-game saying "same number" of wolves
even if there are more wolves than villagers, hunter/vigilante dying
during night sometimes not clearing variables correctly (thus causing
premature night end) and some various stupidity going on with some old
code I wrote that doesn't have any visible effects.
* Combine all players into the same succubus death message
* Fix stylistic issues and succubus idling not working
This doesn't show the variables for a few specific functions, specifically those which have a '_ignore_locals_' local variable set to a truthy value. For now, those functions are handle_error.__call__ (which is present in every traceback ever, due to its nature), cmd.caller and command.caller (one or the other is present in a large number of tracebacks, again due to their nature).
This makes debugging easier when the error is not in the innermost frame, but rather in one or two frames before, and the locals are lost.
var.ALL_PLAYERS update (for !swap) needs to happen before checking if the new user is in var.ALL_PLAYERS
pl is inaccurate during nick changes because the user's nick in var.ALL_PLAYERS is updated automatically but in var.ROLES isn't, therefore a user that changes nicks would never be returned in list_players()
one instance of prefix incorrectly changed to user.nick
If it is impossible for other (non-wolf) roles to have killed at night,
we do not deduct from the traitor count. This logic isn't perfect yet,
but should cover the majority of cases.
