Handle non-ISO-calendar edge case with leap years and differencing

[catamorphism]

See #3159

[catamorphism]

test262 PR: tc39/test262#4636

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.67%. Comparing base (1962c8b) to head (dcf62ff).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3176      +/-   ##
==========================================
+ Coverage   96.66%   96.67%   +0.01%     
==========================================
  Files          22       22              
  Lines       10321    10357      +36     
  Branches     1857     1859       +2     
==========================================
+ Hits         9977    10013      +36     
  Misses        294      294              
  Partials       50       50              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[ptomato]

I think this fixes some cases but regresses others.

d1 = new Temporal.PlainDate(2016, 3, 31, 'chinese');
d2 = new Temporal.PlainDate(2019, 3, 29, 'chinese');
d1.until(d2, { largestUnit: 'years' })
  // old: P3Y
  // new: P37M
  // The old result is dubious (looks like it might need to be P3Y1M) but the new result seems wrong in any case

d1 = new Temporal.PlainDate(2019, 5, 1, 'chinese');
d2 = new Temporal.PlainDate(2020, 4, 19, 'chinese');
d1.until(d2, { largestUnit: 'years' })
  // old: P1Y
  // new: P12M
  // The old result seems correct (Chinese year 2019 is a common year)

I suspect this is because the assumption mentioned in the comment is incorrect: that the difference is less than a year when we hit the problematic case.

I've pushed the in-progress snapshots I've been working with to https://github.com/tc39/proposal-temporal/compare/chinese-hebrew-snapshots?expand=1 so you can do the same testing I'm doing.

[ptomato]

This looks correct to me. I have just a couple of nitpicks.

[ptomato]

Ditto

[ptomato]

Thanks!