Oli

I addressed some of the TODOs and fixes the unit and integration test failures. Also added release notes that cover all new functionality of this PR. I added a "Status"...

> Why don't we just compute the BIP86 value and drop it into that structure so that we can unify the types without needing the Option? I'm trying to push...

Rebased after https://github.com/lightningnetwork/lnd/pull/8520 and addressed some first comments.

I extracted some commits into https://github.com/lightningnetwork/lnd/pull/9025 as requested. All TODOs except for a single one (will just be an additional itest) are addressed, so all parts are now officially ready...

@ProofOfKeags and @yyforyongyu: I believe I have addressed all of your comments to **part 1** and have extracted those commits into https://github.com/lightningnetwork/lnd/pull/9030 as requested. Please do the second round of...

@ProofOfKeags and @bitromortac, I've addressed all your comments on part 2 (except for those being tracked as a TODO in the new PR) and extracted it into a separate PR...

@ellemouton (and @bhandras), I addressed all your comments for part 4.

I didn't get to addressing all comments before my break. I resolved those that are addressed.

While this implementation is likely correct and probably the thing we _should_ do, I worry about the performance hit this will take. Do we have a benchmark test to see...

Nice benchmark tests! An alternative to syncing on every write would be a time based sync. So every X seconds we'd call `Sync()` on the header store's file. Or a...