capture call stack at each event

investigate if we can efficiently capture and record the call stack at each event, which would provide more details about how execution flowed to a particular point in the code.

this could be naively the raw stack memory region, optionally sharing structure across entries for performance (premature optimization?). or, maybe it could just be a list of return addresses recognized from the stack.

No files, tests, or entry points are named in the issue. Start by locating how events are dispatched and recorded, then investigate whether each event can retain the raw stack memory region or a list of recognized return addresses without unacceptable overhead. Done means selecting and documenting a feasible call-stack representation and its performance trade-offs.

Written by the indexing model from the issue text.