Delete leaves from queue in UpdateLeaves

[gdbelvin]

This change brings the mysql implementation into line with what spanner does so we can have a consistent behavior between implementations.

Previous behavior:

• mysql impl deleted leaves from the queue inside Dequeue
• spanner does not delete inside Dequeue, it does it in UpdateLeaves

New / spanner behavior:

• mysql deletes leaves inside of UpdateLeaves

• updated mysql tests

• I have updated the CHANGELOG.

  • Adjust the draft version number according to semantic versioning rules.

• I have updated documentation accordingly (including the feature implementation matrix).

[Martin2112]

Looks like the batched queue version of the code will need to be adjusted as well.

[gdbelvin]

After taking a closer look at the Batched Queue implementation, I'm not sure if I can make this change. The batched queue version takes QueueID, a value not passed in LogLeaf.

I'm unsure how to proceed now, since there's not really a behavior we can standardize and test against between the different storage implementations.

[gdbelvin]

@Martin2112 @AlCutter -- would love your thoughts on this. I was going to abandon this, but the cloudspanner dequeue operations are currently broken, and this is a prerequisite to having dequeue tests.

Update - I think I have a solution

[codecov]

Codecov Report

Merging #2159 into master will increase coverage by 0.21%.
The diff coverage is 76.47%.

@@            Coverage Diff             @@
##           master    #2159      +/-   ##
==========================================
+ Coverage   64.24%   64.46%   +0.21%     
==========================================
  Files         116      116              
  Lines       10002     9919      -83     
==========================================
- Hits         6426     6394      -32     
+ Misses       2958     2921      -37     
+ Partials      618      604      -14     

Impacted Files	Coverage Δ
storage/mysql/queue.go	67.24% <75.00%> (+24.10%)	⬆️
storage/mysql/log_storage.go	68.22% <77.77%> (-0.55%)	⬇️
merkle/log_proofs.go	93.93% <0.00%> (+24.65%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 67ba297...0deeef5. Read the comment docs.

[gdbelvin]

@Martin2112 the build is passing now. Not quite sure why the github / travis hook isn't picking it up.

[gdbelvin]

Tweeked the error message. Anything else needed?

[Martin2112]

As long as we're happy that this is being properly tested and won't cause issues for users.