Resolving remaining concerns with Pin docs #119714

#116129 was landed with some outstanding concerns, since it was considered to be an improvement over the status quo despite those concerns. Would be good to further improve the docs and resolve those concerns. :) The text also changed a bit since the concerns were originally raised, so we have to re-evaluate if they still apply.

1. This thread is about the concern that the description of Unpin focuses too much on describing the effects of !Unpin rather than Unpin.
2. This thread is about the docs basically re-telling the story of how we arrived at Pin, with various twists and turns, which might not be the best way to explain how to work with the final design. The historical part should possibly be moved further down in the docs.
3. This thread is about the interaction of structural pinning and Unpin.
4. This thread is about the claim that Unpin types will "not expose nor rely on any pinning guarantees", which is not quite correct.
5. This thread is about the "pinboard" analogy not being evocative or necessarily very useful.