The central case I want to capture is that of the person who worked to earn, and is still capable of doing so ..., but decides not to ..., and has enough money saved or in a pension to support them for the rest of their life.

Well, that's a reasonable definition. But this particular concept of retirement is not new at all and does not mention retiring in cohorts or at a particular age. I am pretty sure some Romans retired like this a couple of thousand years ago :-)

And one of the causes of particular jobs disappearing is automation.

Since we're talking, basically, economics we might as well use the proper terminology. You are making the argument that any increase in the productivity of labor leads to job losses.

This is useful because it helps us predict whether, in the future, increased automation might quickly eliminate many jobs.

If you want to base your prediction on available evidence (aka historical data), note that other than in short term, increases in the productivity of labor have not led to massive unemployment.

There are processes driving job loss and separate processes driving job creation.

No, I don't think they are separate processes, I think they're much more like flip sides of the same coin. You can't create without destroying. For someone to become a scribe, he has to be not needed as a farmhand.

In general, when thinking about jobs, you might want to think in terms of "creating value" and "(re)distributing value". A job is not a benefit -- it's a cost (in human time and effort) of producing value.