The motor casings should not be live to any part of the electric circuit, unless it was thought a good idea to connect them to battery -ve. Some very old motors did have one brush insulated from the case and the other using the case as a connection, but all of the more modern motors that I have seen (modern = after 1960) have both brush terminals insulated. I have come across motors containing bits of swarf that allowed one of the brushes to contact the metal frame, but that was not a design feature.
There might be a problem with the ground line wiring, a bad -ve power connection to an ESC causing it to get that connection through the thin radio lead does give all manner of strange effects, up to and including vapourising the metal inside the black wire of the radio leads, but on its way there, the fluctuating ground reference voltage wreaks all sorts of misoperation.
No familiarity with the particular transmitter, but if it is capable of mixing, it is possible that an operators thumb "might", while moving it up/down, introduce a bit of unintended sideways. Acid test - move the right stick left/right, does anything but rudder movement happen?