Mystic the action

[lbellini]

 4G^   1G^

 3B>   4B<

 2B>   1B<

 3GV   2GV

 3:   MYSTIC the action

                 .     .     .     .    4G^   1BV   2B^   1GV

 3G^   4BV   3B^   2GV    .     .     .     .

sd has this ending in a 2x8, but i would expect a parallelogram with head corners push casting and side corners crashing.

Note that this is the first bug opened due to my shiny new (partially working) parser. There are many more, but since this one actively does the wrong thing, I'm opening a bug.