Fix bitcoind shutdown hang

[benthecarman]

Previously, the shutdown process could hang when using the bitcoind
backend because the syncing process was not always checking if we had
received the shutdown signal. Now with any future we call during the
sync process we will use a `tokio::select!` to make sure we abort early
if we receive a stop signal.

[ldk-reviews-bot]

👋 Thanks for assigning @tnull as a reviewer!
I'll wait for their review and will help manage the review process.
Once they submit their review, I'll check if a second reviewer would be helpful.

[tnull]

So, the changes to bitcoind behavior here make partially sense to me, as we want to respect the shutdown signal when retrying the initial sync.

However, for the other changes I'm not so sure: a) we generally should attempt for chain syncing to finish gracefully, and picked corresponding timeouts to at least try. b) For Electrum we use a bunch of spawn_blocking internally, which can't be aborted. This means that even though we'd abort polling them, Electrum would still make changes to our LDK objects, which could lead to unexpected behavior if we had already shut down our background processor and hence won't re-persist.

Could you check whether the issue you were seeing would be accounted for by just the changes for the initial bitcoind sync? If not, we'll have to further debug why the timeouts aren't respected.

[benthecarman]

Yeah did not need the first commit. Fixed up based on your comments for the second commit.

[tnull]

LGTM

Please rewrite the commit message to better state why this change is necessary and what it does (preferably in 1-2 paragraphs max, nobody will want to read a book of ~inaccurate AI slop in git log).

[tnull]

Landing, CI failure should be fixed by #689.