Even if you did short the 4 and 8 ohm taps together it would just present a very low impedance for a brief interval. That’s not as dangerous as it would be with a transistor amp, and far less so than allowing the voltage to spike by going open. Overdissipating a pair of output tubes for less than a second is not harmful at all. And that would really only happen at war volume.
I do have a switch on one amplifier, Tubelab SSE, but it is just a simple DPDT to alternate between the two secondaries. I have changed it in the last 6 months, just to check if a slightly 'strident' sounding piece of music was the same on the other output impedance. The switch is tucked out of the way and normally I toggle it when the amp is off.
The problem is that anything too advanced just adds complexity where it is not really that critical. For instance if you have NFB from the secondary, is the wiring for that affected? What is the quality of the 2 DPDT switches you propose, and where to drill the holes? An alternative is an just to add an extra binding post on each channel, connected to the other secondary tap. Adding the 200R resistor across the secondary permanently is an insurance policy, something I should always do in the future.
The problem is that anything too advanced just adds complexity where it is not really that critical. For instance if you have NFB from the secondary, is the wiring for that affected? What is the quality of the 2 DPDT switches you propose, and where to drill the holes? An alternative is an just to add an extra binding post on each channel, connected to the other secondary tap. Adding the 200R resistor across the secondary permanently is an insurance policy, something I should always do in the future.