This'll take quite a while, but there are no shortcuts!
Whilst connected to the dim-bulb tester (a slip of the probe can be very expensive) and with 'AUTO CLASS A' OFF/disabled, measure, compare and tabulate the BASE, EMITTER & COLLECTOR voltages (with reference to 0V) as follows;
TR223 & TR224
TR225 & TR226
TR227 & TR228
If there are no obvious discrepancies between the L & R channel results, perform the same measurements with 'AUTO CLASS A' ON/enabled.
So before I do what you said, I ran my own flavour of testing which is with everything off and comparing B-E resistances (bit safer), and I've spotted that the B-E resistance of TR228 (bad channel) is 101kΩ when it's 1.5-2.1MΩ on TR227 (good channel).
Something tells me TR228 or something near it (likely D218) is bad.
Judging by the in-circuit continuity test on D218 vs D217 I'm betting the diode is the culprit!
I've ordered the remaining caps to finish her off, the owner will be thrilled.
If anyone knows where to buy a rec-out knob (grubscrew type) let me know too.
If the knob is the same type as the AX-700, with 18mm OD and 6mm hole for the shaft, I haven't been able to find an exact sub. But I was able to find and adapt a reasonable temp solution until I find a better one...