Regarding that current sensing in the ground connection, shame on Rigol. Every serious article about power supply design tells you not to do it on the ground connection, unless the channel is completely isolated. With that current sensing their claim to a precision power supply is just hogwash.
Except putting the sense resistors low here has caused no problem, and nor would I expect it to. It'd work just fine if they strongly tied the two common terminals together, and just had one voltage sense wire sampling the potential at this "one true ground". The thing that I find really interesting and original (to my newbie eyes) about this schematic is that CH2 and CH3 are completely isolated, except for the one connected ground sense point (let's just pretend they commoned them properly, not via sense wires). In particular, they have completely separate coils on the main transformer, separate rectifiers, etc etc, all the way out to the outputs,
and all the way back again. That makes the location of the current sense resistors completely irrelevant according to your basic Kirchoff's Voltage Law. Rigol just put them low because it's easier to measure there.
I'm a bit tired and what I wrote above might not make sense, but let's at least agree on this point:
if they moved the current sense resistors to the high side, but left the messed up commoning-via-sense-wires, then the voltage error with "incorrect connections" would still exist. Nothing to do with that problem at all.