[18esp] fix mz mza cheater token

[bentziaxl]

MZA and MZ share a single slot city in madrid (f24).
When mz is taken over, and MZA was placed as a cheater token, the closing of MZ doesn't remove the extra slot.
added minor code to fix this behaviour.

[ollybh]

I think that this might not work in all cases.

Would it be possible to do something like this, overriding close_corporation instead of adding the fix_cheater_slot_madrid method?

def close_corporation(corporation, quiet: false)
  if corporation.type == :minor
    corporation.placed_tokens.each do |token|
      city = token.city
      remove_slot = city.tokens.any? { |t| t.cheater? }
      city.delete_token!(token, remove_slot: remove_slot)
    end
  end
  super
end

[bentziaxl]

I think that this might not work in all cases.

Would it be possible to do something like this, overriding close_corporation instead of adding the fix_cheater_slot_madrid method?

def close_corporation(corporation, quiet: false)
  if corporation.type == :minor
    corporation.placed_tokens.each do |token|
      city = token.city
      remove_slot = city.tokens.any? { |t| t.cheater? }
      city.delete_token!(token, remove_slot: remove_slot)
    end
  end
  super
end

there is no cheater? method, the best i could come up with is city.tokens.size > city.normal_slots
as cheater adds an extra token, but not another normal_slots.

[ollybh]

there is no cheater? method, the best i could come up with is city.tokens.size > city.normal_slots as cheater adds an extra token, but not another normal_slots.

My mistake, it should have been token.cheater (without a question mark), not token.cheater?.

[bentziaxl]

found 2 new minor bugs that i'm adding to this:
1- @game.city_by_id('F24-0-2') breaks when the city has been upgraded to green (so obvious!)
added code to determine the city programatically.

2- the rules say that the first run on 2 permanent needs to connect both aranjuez and madrid, not just aranzuez. fixed the logic to reflect that.

3- change the code to token.cheater as suggested,

[ollybh]

@bentziaxl Once a review has been started, can you avoid force-pushing to the PR branch? It means the reviewer can't see what has been changed in the newer commits.