bump(bdk_chain): rusqlite to 0.36.0

[binarybaron]

Description

This PR bumps the chain crates feature dependency on rusqlite to the more recent version 0.36.0.

Notes to the reviewers

I submitted this PR because the "outdated" dependency breaks dependency resolution in my project.

See #1766

Changelog notice

Checklists

All Submissions:

• I followed the contribution guidelines

New Features:

• I've added tests for the new feature
• I've added docs for the new feature

Bugfixes:

• This pull request breaks the existing API
• I've added tests to reproduce the issue which are now passing
• I'm linking the issue being fixed by this PR

[tnull]

I don't think that unilaterally bumping the rusqlite dependency just in BDK is a good idea, as it will result in incompatibility with other crates in the rust-bitcoin space. Unfortunately, 0.31 is the current 'agreed upon' standard across crates.

If we see a need to bump it, this will take some coordination effort across projects (BDK, LDK, Cashu, Fedimint, Payjoin, at the very least would be affected).

[binarybaron]

Changed to 0.36.0 because arti-client uses 0.36.0

[notmandatory]

@tnull & @binarybaron what would you think about extracting the bdk_sqlite persistence store from the bdk_chain crate and putting it in it's own crate. If we do that we could have a 1.0.0 version with the current 0.31.0 rusqlite and release a new 2.0.0 version with 0.36.0 rusqlite, while still back-porting bug / features for the 1.x. Or open to other ideas along these lines.

[tnull]

@tnull & @binarybaron what would you think about extracting the bdk_sqlite persistence store from the bdk_chain crate and putting it in it's own crate. If we do that we could have a 1.0.0 version with the current 0.31.0 rusqlite and release a new 2.0.0 version with 0.36.0 rusqlite, while still back-porting bug / features for the 1.x. Or open to other ideas along these lines.

So, moving the SQLite store into an individual crate would generally be a good idea, IMO. I however don't think that it would solve the issue, at least if you don't want to go down the road of introducing features to enable individual rusqlite versions. As mentioned above, a lot of projects in the space by now depend on the exact rusqlite version, and I think it would be important to check in with most of them to coordinate such breakage. Apart from that, yes, it would of course also require a major version bump for all of the involved projects.