The delta calculator's doc names the census that actually guards the still-naked delta-family list (#3540 follow-up)

[erikdarlingdata]

One comment, corrected

#3595 (V127 / Lite v60, #3540) rewrote CollectorDeltaCalculator's "every consumer maps 0 to NULL" claim so it names which delta families persist sample_interval_seconds and which four still do not, and pointed readers at "a delta-family census in Lite.Tests (SharedCollectorDefaultsPinTests)" as the guard that keeps the naked list shrinking deliberately. The census is real, but it lives in Lite.Tests/DeltaFamilyIntervalColumnTests (StillNaked + EveryDeltaFamily_PersistsItsInterval_OrIsNamedAsStillNaked) - the pointer named the file I first planned to put it in. Comments here are how the next contributor navigates, so the class name is fixed to the one that actually fails when a seventh family ships naked.

Raised as a non-blocking nit by the bot review on #3595; auto-merge had already fired, hence the follow-up rather than a push to that branch. Comment-only; no behaviour, no test changes. Builds with 0 warnings.

Which component(s) does this affect?

• Lite
• Darling
• Lite Tests
• Darling Tests
• SQL collection scripts
• Documentation
• Full Dashboard (deprecated)
• CLI Installer (deprecated)

Checklist

• I have read the contributing guide
• My code builds with zero warnings (dotnet build -c Debug)
• I have tested my changes against at least one SQL Server version - comment-only change, nothing to run
• I have not introduced any hardcoded credentials or server names

[claude]

LGTM — verified this is a comment-only 1-line fix correcting a stale class-name reference in CollectorDeltaCalculator.cs. Confirmed the new name (DeltaFamilyIntervalColumnTests) and the StillNaked census/test it points to actually exist in Lite.Tests/DeltaFamilyIntervalColumnTests.cs. No executable code, T-SQL, or Lite/Darling parity impact.