This may or may not be relevant, and it's a stretch.
Back in England in the mid-60s I got a new company car, a Ford Cortina. After a couple of weeks I found the same thing; I could turn the ignition off and pocket the key, and the engine would continue to run. I had to place it in top gear and let in the clutch to stall it. I then found that the engine would stop when I turned the lights off. Duh?
I called the local Ford dealership and spoke to the service manager, who told me that it was impossible. "OK," I said between gritted teeth, "I'll drive it down and you look at it." Which I did, causing consternation in the service department, with everyone from the old hands to just-trained apprentices shaking their heads and muttering "impossible." They replaced the ignition and light switches - no help.
Then the dealer management got in a tizzy and called Ford in Dagenham, as it might have been a design fault in the ignition switch (the opposite of what's happening at GM.)
They gave me a loaner, and called one of their electrical experts in from London, who spotted the problem at once. During assembly (as you may remember, English mass production cars like British Leyland and Ford were put together by gorillas hating their jobs and looking for a chance to strike) someone had managed to cross wires in a rear light cluster, and left too much bare uninsulated wire, so that the rear and brake lights finally touched.
Unknown to me the brake lights had been coming on whenever I turned the main lights on; I couldn't know because I couldn't see them. Now the brake light switch was only supposed to be "hot" when the ignition was on, so when the lights were on the current back-flowed at the crossed wires, through the brake light circuit and through the "ignition only" terminal into the coil.