I would think one of the pilots would stand behind the plane and verify control surfaces via radio, but I guess that's too time consuming or error prone?

Works fine for cars though. Drive through lube shops and states with safety inspections will perform indicator checks this way. If you push the left blinker and the right light starts flashing you've got a problem...

Hopefully by now the cockpit has access to external cameras to visualize what's happening behind them in real time

as a pilot, I have to really force myself to 'see' that the aileron is moving the correct direction. Not because it's hard, but because my lizard brain wants so badly to skip over the thing that it has seen so many times that it's no longer interesting.

I have not seen a glider where it is physically possible to cross-connect the controls simply through the process of de-rigging and rigging (removing and replacing the wings and stabilizer, which is often done for transport or storage), though I imagine there were examples in the past. I have, however, seen ones where the controls can be left unconnected or even insecurely connected (though such designs now seem deprecated), in which case they may move in the correct direction on the ground, but fail to do so under load (and may flutter). On account of these possibilities, the visual inspection and pull/twist test of the connections, together with positive control checks (where someone holds the surface to provide resistance), as mentioned in the article you link to, should be SOP.