Okay, so I got the pots in the correct places. I also found that R57 on the attack/release board was the wrong value by several orders of magnitude, so I put in the correct value. I went through the entire circuit and triple checked my resistor color codes and other component values, and everything else checks out, including ratio and meter boards and now the attack/release board.
This mostly fixed the issue. However...
In step 3 of calibration (where I was having trouble), turning on the attack pot does drop the voltage from 2.44v now that I've corrected those components, but not all the way back down to 0.775 -- only to 1.025. If I adjust that down with the output control, then the "attack off" voltage drops accordingly. In other words, I'm not getting the full 10dBu attenuation.
Additionally, my voltages at TP22, Q7-10, and CR2 went further out of whack. Most everything there shot high, particularly at the higher ratios (which it was already doing, but more so now -- for example, I've got 0.450v at Q7 as opposed to the expected 0.377v). I've also still got a higher-than-expected voltage at Q5.
Having said all that, the unit does pass audio and compresses, and on a fun little test running bass guitar through it and then some drums, it sounds fabulous! I'm just wondering what I'm still missing to get it all the way to where it's supposed to be.