pickfirstleaf: fix bug in address de-duplication

[arjan-bal]

Due to a bug in the new pickfirst balancer, it wasn't de-duplicating addresses in the resolver update. The only user visible impact of this seems to be less frequent picker updates after the first pass in happy eyeballs and incorrect interleaving of IPv4/IPv6 addresses during the first happy eyeballs pass.

RELEASE NOTES:

• balancer/pickfirst: Fix a bug where duplicate addresses were not being ignored as intended.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 81.97%. Comparing base (bb71072) to head (fb56e8e).
⚠️ Report is 10 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #8611      +/-   ##
==========================================
- Coverage   82.00%   81.97%   -0.03%     
==========================================
  Files         415      415              
  Lines       40697    40698       +1     
==========================================
- Hits        33372    33364       -8     
- Misses       5937     5945       +8     
- Partials     1388     1389       +1     

Files with missing lines	Coverage Δ
balancer/pickfirst/pickfirstleaf/pickfirstleaf.go	88.08% <100.00%> (+1.00%)	⬆️

... and 18 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[easwars]

This part of the description is very confusing:
The only user visible impact of this seems to be more less frequent picker updates

[easwars]

The release note doesn't seem to indicate that we are actually fixing a bug. It looks more like we are adding the logic to dedup addresses. Although technically this PR is doing that since we weren't deduping before :)

[arjan-bal]

This part of the description is very confusing: The only user visible impact of this seems to be more less frequent picker updates

Fixed. It should have been "less frequent picker updates" since the address list would be longer. I changed the phrasing of the sentence midway through typing it.

The release note doesn't seem to indicate that we are actually fixing a bug. It looks more like we are adding the logic to dedup addresses. Although technically this PR is doing that since we weren't deduping before :)

Updated the release notes to mention that not ignoring duplicate addresses was a bug and it is being fixed.