Upgrade to ring 0.17

[thomaseizinger]

Reboot of #98.

This PR is a breaking change because the RcgenError::RingKeyRejected variant is changed AND because ring is part of the public API via the From impl for RcgenError.

[thomaseizinger]

All commits pass clippy and formatting. Let me know if you want any of these to be split out into other PRs!

[iamjpotts]

See discussion in #98 - was that drawn to a consensus wrt the additional parameter?

[djc]

(FWIW, you had a nice commit history before but now there's a bunch of back and forth there... maybe use rebase -i to skip the commits that introduce the Cow and introduce ring to the public API, instead of reverting them with later commits?)

[thomaseizinger]

(FWIW, you had a nice commit history before but now there's a bunch of back and forth there... maybe use rebase -i to skip the commits that introduce the Cow and introduce ring to the public API, instead of reverting them with later commits?)

Sure! I did see that you are not squash-merging in here so I figured you'd care about the commit history :)

[codecov]

Codecov Report

Merging #166 (dc38152) into main (aaf4dd7) will decrease coverage by 0.04%.
The diff coverage is 76.92%.

@@            Coverage Diff             @@
##             main     #166      +/-   ##
==========================================
- Coverage   72.14%   72.11%   -0.04%     
==========================================
  Files           7        7              
  Lines        1881     1886       +5     
==========================================
+ Hits         1357     1360       +3     
- Misses        524      526       +2     

Files	Coverage Δ
src/lib.rs	75.40% <ø> (ø)
src/error.rs	1.92% <0.00%> (ø)
src/key_pair.rs	72.00% <83.33%> (-0.28%)	⬇️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[thomaseizinger]

Sorry about the closing stuff, I fuffed around with rebase too much 😅

This PR is now entirely focused on upgrading ring.

[est31]

Okay except for the two comments by djc and cpu, where the one by djc can also be addressed in a subsequent PR (as long as it's done before a release).

[thomaseizinger]

Okay except for the two comments by djc and cpu, where the one by djc can also be addressed in a subsequent PR (as long as it's done before a release).

Fixed! It is a breaking change so you might want to release a non-breaking one before merging this one.