Document the handoff to nomologR on the README, and fix the RC-day order

[JUhalt]

The maintainer asked that each package's README document the handoff in its own section near the bottom. nomologR is adding the mirror section, "The handoff from contentvalidR", above its "Stability" section. Docs only: no code, no computed value, and no handoff change.

"The handoff to nomologR", just above References

• What a handoff carries:
  • the carried and held-back items, each with its status, recommendation, and rule;
  • the scales;
  • each item's statistics, intervals, and criteria, and the panel's agreement;
  • two facts only the analyst knows: reverse_keyed (or character(0) if checked and none is) and response_scale.
• Making one: a runnable content_handoff() call on the README's own first analysis. It shows the two carried items.
• Passing it to nomologR: nomo_screen(responses, items = h) and nomo_run(responses, scales = h), shown but not run. Pass the whole handoff, not h$items.
• Reading a handoff: the reader contract in ?content_handoff (State the handoff reader contract nomologR asked for, and record the joint 1.0 checklist #79).
  • Decisions come from carried and status, never re-derived from a statistic against its criterion or from the producer version.
  • recommendation is prose.
  • keying of NA means unknown.
  • nomologR uses the same wording in its section.
• Versions: schema 1 has been frozen since 0.7.0, with additions only at the end. A version 2 would be produced beside version 1 for at least a release cycle. The nomologR reader is tested against producers 0.6.0 through 0.9.0, with 0.10.0 fixtures delivered today. The section links ?content_handoff, the handoff vignette, and the joint walkthrough.

Also

• ROADMAP.md: release-candidate day on 2026-10-17 now runs in the order agreed with nomologR, so that nomologR's tests pass before either package tags:
  1. The 0.99.0 commit and its fixtures come first, with the commit recorded in the manifest.
  2. Then nomologR's suite runs.
  3. Then both packages tag, contentvalidR on exactly the commit the fixtures came from.
• Version: opens 0.10.0.9000, with the development NEWS heading.

Verification

• The README knits. The handoff chunk runs and prints "Clear 1" "Clear 2".
• Spelling is clean.
• tools/release-gate.R on this PR's head (be3097d), run in a detached worktree: build, smoke, and check all pass; R CMD check 0 errors, 0 warnings, 0 notes (the usual CRAN-incoming and V8 notes did not appear in this run). A first run failed only because repository.upenn.edu returned a transient 500 for the Krippendorff (2011) link; it answered 200 minutes later, and the rerun on the same commit passed.

🤖 Generated with Claude Code