refactor wallet address caching into its own public method

[a5an0]

Description

Currently, the only way to ensure that a wallet's internal database has addresses loaded and cached is through Wallet::sync: that function generates and caches up to a number of addresses if they aren't already in the database, and then uses the wallet's blockchain client to sync those addresses. If you are using an offline wallet, there is no mechanism to ensure that the database has addresses loaded. This is a problem for usecases like an offline wallet being used as a multisig signer and wanting to validate change addresses as Wallet::is_mine will only work properly if the owned-address is loaded in the database.

This PR takes the address caching functionality out of Wallet::sync and puts it in a new public method that is available to offline wallets. Wallet::sync uses this method internally.

Checklists

All Submissions:

• I've signed all my commits
• I followed the contribution guidelines
• I ran cargo fmt and cargo clippy before committing
• I've added docs for the new feature
• I've updated CHANGELOG.md

[LLFourn]

Concept ACK. Like @a5an0 mentions, we need this when you give your descriptors over to another machine which generates addresses without your knowledge and then hands you a PSBT to sign. HW wallets like coldcard do this kind of thing too.

An important side note is that caching addresses is not even technically necessary for sync I think. You can just generate the new ones on the fly (up until stop_gap) as you look them up. Right now our sync logic depends on pre-caching but I want to fix this.

[afilini]

Try rebasing on the lastest master, it should fix the CI

[afilini]

Concept ACK btw, this seems useful.

Right now our sync logic depends on pre-caching but I want to fix this.

Yes, this would be great. The current sync/setup behavior is also very inconsistent, we should think about what to do there.

[notmandatory]

Once #502 is merged you'll need to rebase to pickup some CI changes that are in that PR. Then this one looks good to go.

[a5an0]

rebased onto master to pick up CI changes

[LLFourn]

This LGTM just a few cosmetic things that I think we should fix up.

[a5an0]

Thanks for the feedback! updated.

[LLFourn]

ACK 66c6158 squash and merge

[notmandatory]

ACK edf2f0c