First off - great that you opened up the unit and tested disconnecting and reconnecting cables - that goes a long way to figuring out the issue (or at least eliminating some things). It is a bit of a "head-scratcher", though! If it was a true short circuit, then an entire row or column of the keyboard matrix wouldn't work (e.g, it would affect several keys), unless that specific row and column were somehow shorted out, or if that specific key (at the end of the left board) just happened to have its own row and column by virtue of being 'outside' the matrix group. I kind of doubt it - there are 24 keys on that board, so the likely groupings are 4x6 or 3x8, which means that there are 3 or 4 controls lines and no 'outlier' key. But it's not as if the key doesn't work at all - it's producing an initial attack, then somehow repeating with a string vibrato kind of sound, which (to me) means that the decoder is at least 'seeing' the keypress. It's not likely to be the diode either, or it wouldn't make a sound at all.
If you hold the "bad" key down and press all other keys, do the other keys work properly? Can you play the bad key soft and loud (i.e., does it respond to velocity)?
My initial gut feeling is that it's software-related or the decoder chip is somehow bad as opposed to a short-circuit, but those are just guesses. Maybe try the two suggestions and let the group know what you discover.
