Comment # 28 on bug 1205111 from Takashi Iwai

(In reply to Wolfgang Engel from comment #25)
> Kernel 6.0.7-4.ge0a2b69 works!

Good to hear!

The problem was two folds: the device declares the third interface as a MIDI
interface, but the driver couldn't probe it because the descriptor is
vendor-specific and incomplete.  Meanwhile, there was a bug in the driver that
skips the card registration if the interface probe fails, and since the failed
interface is the last one, the whole card registration is skipped.

I'm going to submit the fixes to the upstream and backport to TW and other
kernels.

So the last question is: does this device have really a MIDI interface? 
Through a quick glance at web pages, there seems no (physical) MIDI I/O.  If
there is any, I can provide a patch to forcibly enable the interface, too.