Fix a 0.2% natschannel_test flake with some forced scheduler sleeps

[evankanderson]

Summary

I noticed natschannel_test failing in a CI run (which seem to run on a slower CPU than my MacBook). I was able to reproduce with -count 1000 with some background adversarial tasks. Putting in a forced scheduler transition (aka time.Sleep) after closing the subscriber seemed to prevent losing the third message (it was always the third message, which made me suspect the lines just before that publish).

Change Type

Mark the type of change your PR introduces:

• Bug fix (resolves an issue without affecting existing features)
• Feature (adds new functionality without breaking changes)
• Breaking change (may impact existing functionalities or require documentation updates)
• Documentation (updates or additions to documentation)
• Refactoring or test improvements (no bug fixes or new functionality)

Testing

go test -count 1000 ./internal/events/nats while running go test -count 2 ./... in parallel in another window.

Review Checklist:

• Reviewed my own code for quality and clarity.
• Added comments to complex or tricky code sections.
• Updated any affected documentation.
• Included tests that validate the fix or feature.
• Checked that related changes are merged.

[coveralls]

Pull Request Test Coverage Report for Build 11583782223

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.01%) to 53.042%

---

Totals
Change from base Build 11579859014:	0.01%
Covered Lines:	13364
Relevant Lines:	25195

---

💛 - Coveralls