clarify interaction between `panic = abort` and backtrace capture

In our project at work we've been seeing issues with backtrace's not capturing the full context of the application state when we panic, where as the builtin backtrace printer for the default provided panic hook has no such issues.

https://github.com/ZcashFoundation/zebra/issues/1471 shows an example of a backtrace that gets cut off, it only shows the frames pushed onto the stack after calling expect, but doesn't include any of our application code's frames otherwise, where as the metadata of the issue captured the location the panic was created from via #[track_caller] just fine.

I'm not convinced I understand what is going on here, or whether this is a bug or expected behaviour, disabling panic = abort fixes the issue though, at least in the playground.

Start by reproducing the difference between panic = abort and the default panic hook in the playground, using a panic from expect and observing the captured frames. Compare the backtrace output with the panic location reported through #[track_caller]. Done means documenting whether the truncated application frames are expected behavior or a bug, and clarifying the interaction.