there were problems with stdio, this is just ok for now
some help from Nikola for some bugs

i(alexander) wasn't sure if it can work well bidirectionally,
but it seems it's ok

…d_location for rr

…or Replay

loading locals still TODO in rr; overally not very tested in db

… not write

…ete generated test

tcp test by mistake committed before: generated with gemini/zed iirc: removing,
maybe if we add full tcp support we can add it

problem: dap backend server still seems to have an error, at least stdio one
works

* testing:
  * add a simple test for it in rr_test
  * add a timeout for the rr test for now with the `ntest` crate
* impl:
  * start refactoring flow_preloader; not finished

… for location

for now working in simple cases for db-backend, and at least not crashing for
some for rr; but not yet a good example of working for rr;

a small fix that tries to repeat the function first/last info for each of the
lines in range in expr_loader; maybe we can instead check the region but for now
hopefully an ok fix

IMPORTANT: some problems overally: jumps and some flow in db-backend seem to be
broken

not sure why we seem to still receive additional stopped after maybe
complete-move/response, but it doesn't seem to be a problem

i forgot that now we mostly move with `self.replay`, so just changing the
handler step_id might not be enough

problem with iterate_asteroids flow remains

also TODO: add rr logic for the new primitives

moved setBreakpoints logic to handler; implemented some of the methods
and placeholders for the others

…re jump_to_call

TODO: rr jump_to_call

…he newer value form

useful for rr: for now add some convert functions for compatibility;
maybe in the future try to replace the old `Value` with the newer versions

try with upstream `rr` in PATH and ct rr worker trying to use it:
but hanging; the test seems to be ok with the dev worker

building but not tested correctly: also more work! TODO