I'm not (yet) an AOMCI member, you guys don't know me from Adam. So feel free to close ranks & laugh behind my back. It's ok. Seriously!
The harness wires to the switch ohm'd out as I suspected: from wire to wire: infinity with the switch open, 0.2 ohms when closed, and neither lead had any communication with ground in either switch position.
So I reconnected the harness wires to the magneto plate's kill wires, but left the vacuum switch out of the equation (minimizing variables). Both plug wires spark with the switch open, and both are dead with kill engaged. OK....getting somewhere.
So I wired the vacuum switch back in. Really don't see how it can have an effect based on what it is & what it does, but ok. Guess what? Both plug wires spark with the switch open, and both are dead with kill engaged.....just like they're supposed to!
I cannot duplicate the problem. Can't.
So I'm sure it looks like the only variable in the equation would seem to be me. Frankenstein engine, non-OMC wiring, home-made harness, modern Deutsch connectors, etc. "Operator error"...."Nut behind the wheel". Or as they say in the IT industry, "I think it's a ID-10T error". I hear ya. But I didn't imagine the spark. Really, I didn't. And I haven't changed a thing. Just disassembled & reassembled. Twice.
If it works, it works...and I shouldn't overthink this. But I saw the spark on the 'top' wire with the kill engaged. I really did. I'll do some more tinkering tomorrow, but for now everything is exactly as it should be. I have no explanation.
I appreciate all of your replies, and I'm sorry there's no "eureka moment".