Don Wilhelm W3FPR gave the me answer for fixing my K2/100. He suggested I replace D16 and D17, as they may have gone bad without failing open or shorted.
It took a while to get the parts from Mouser. But these 1N5711 diodes were only $0.10 apiece. So, I bought 10 in order to have some spares.
Replacing D16 and D17 did the trick. I managed to test one of the diodes I removed (the other was destroyed in the desoldering process), and it showed a nearly correct forward voltage on my DVM, and about 3 times that value when connected in reverse. The diode action of the part was clearly compromised. A new part gave a low forward voltage, and an open when connected in reverse.
I suspect these diodes, and possibly the EL5146C might have been damaged by a stray static discharge sometime in the last year. Perhaps a good thing to have on hand -- although Don did indicate that 1N4148 diodes would work OK as a temporary substitute.
With the new diodes in place, the SWR Null Adjustment went well, giving a clear null down below 100 mV. More importantly, it read an SWR of 1.0 on all bands.
The rest of the alignment went smoothly, and the rig appears to be producing the correct power levels on all bands. Made a quick contact with HA8RM on 80m to confirm the rig is working again. I think it is cured.