When the switch closes, the capacitance is discharged. The only things limiting the current is the ESR of the capacitor and the resistance of the switch, so the peak current could easily be quite large. However would 10 nF or 100 nF actually hold enough energy to damage the switch?
I normally use a fairly low value resistor in series with the cap, so there is never excess current through the switch. It may not be really necessary, but I just don't like to take chances with reducing the switch life.
You may not see a spark in the dark if it is low energy but if contact area is tiny and gold plating is < 100 nm it might be UV emission with enough heat to erode the gold. Apple lightning connectors are famous for this mode of failure in 1 yr.