* fixed wrong manually pay by mana pool (it pays generic cost instead colored part of monohybrid);
* fixed not working cost reduction effects (now monohybrid cost will be reduced correctly with some limitation, see #6130);
* Added support of spell ability choice in free cast (chooseSpellAbilityForCast);
* Added support of "up to" targets setup in addTargetAmount (use TestPlayer.TARGET_SKIP to stop target choose);
* Restored multiple checks for addTargetAmount;
* Fixed many wrong tests (wrong order or missing commands);
* Gain control abilities - fixed wrong target chooses by AI (selects weakest permanent instead most powerful);
* Target of an opponent’s choice abilities - fixed that AI was able to cancel card cast, fixed wrong target chooses (Evangelize, Echo Chamber, Arena, Preacher, etc);
(First time contributing so please excuse my noobness) Card not prevented by an effect should be playable ("&& !prevented"), but currently AI says card not prevented is NOT playable ("&& prevented"). Also, not sure why AI doesn't add to unplayable even though its not contained by playableInstants or playableNonInstants. I'm not even sure if we're maintaining ComputerPlayer's decision logic (as opposed to Computer Player7).
* Fixed that computer was able to target own permanents for bad effects and opponent permanents for good effect;
* Improved target choices on card score (as example: boost most valuable, discard most weaker);
* Tests: fixed false positive tests (AI depended);
* Tests: fixed that test player can't find permanents for some filters/predicates;
Replaced all instances of "her or she", "him or her", and "his or her", with "they", "them", and "their", or sometimes "that player". Updated verbs where appropriate.
* added support to search targets from multiple opponents instead one;
* added support of TargetCreaturePermanentAmount and all new target types;
* added new priorities (kills most valuable first);
* added bad and good conditions (choose most valuable in good case and most weakest in bad case, e.g. damage yourself);
* fixed that AI targets opponent for good ability instead yourself;
draftaholics data was very skewed towards masterpieces. which meant
decent cards were ranking very low relatively
I used the relative ranking from draftaholics and generated a normal
distribution mean60 std 20 for these rankings to smooth it out.
