fix(ci): code coverage actions/upload-artifact version

[oleonardolima]

Description

The Code Coverage step seems to start failing due to actions/checkout@v2 becoming deprecated, and an error now instead of a warning. At least it's what I got from the error message here: https://github.com/bitcoindevkit/rust-esplora-client/actions/runs/10702593999/job/29671147653?pr=93

This PR bumps it's version to v4, the latest one. I'm also taking the opportunity to bump the action/checkout to v4 too.

I'm not sure if it became fully/enforced deprecation from yesterday to today, because it ran successfully on the last master merged PR CI steps #:thinking:

Notes to the reviewers

Please let me know if I should take the opportunity to bump any other actions, or update the workflow in any way.

Changelog notice

• Bump actions/upload-artifact to v4.
• Bump actions/checkout to v4.

Checklists

All Submissions:

• I've signed all my commits
• I followed the contribution guidelines
• I ran cargo fmt and cargo clippy before committing

[ValuedMammal]

This looks good to me, seems like the workflow was in need of an update. I noticed it's also installing rust via sh.rustup.rs which is interesting considering we typically use another github action for that

[oleonardolima]

This looks good to me, seems like the workflow was in need of an update. I noticed it's also installing rust via sh.rustup.rs which is interesting considering we typically use another github action for that

Oh, you're right. I'll update it to the same one as the other CI jobs.

[coveralls]

Pull Request Test Coverage Report for Build 10723811847

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 87.438%

---

Totals
Change from base Build 10690950735:	0.0%
Covered Lines:	1058
Relevant Lines:	1210

---

💛 - Coveralls

[ValuedMammal]

ACK 8608109

[tnull]

ACK, mod one comment

[tnull]

Why is this change necessary? IIUC, it should do the same thing, but introduces another (unnecessary) dependency to the CI? IMO, the previous way was preferable (and saves the addtional rustup update step later on).

[oleonardolima]

It's not really necessary. I just used the opportunity to update it to use the same toolchain action used on our other CIs.

I have no big feelings about dropping the e404089 if you folks see fit :)

[ValuedMammal]

I have no big feelings about dropping the e404089 if you folks see fit :)

Sounds good, I guess tnull is right

[oleonardolima]

Awesome, I removed the e404089 commit and rebased it on top of master.

[oleonardolima]

I added a new commit 7321a5d to fix the tokio-util MSRV problem.

[evanlinjin]

ACK 7321a5d

[ValuedMammal]

ACK 7321a5d