docs: bring the v1.14.0 notes up to what has merged

[seabbs-bot]

v1.14.0 is still untagged, and its notes stopped at the release-prep pull request. Eleven have merged since, several with no entry at all. I audited the notes against every commit in v1.13.2..main rather than against the pull request list.

What was wrong

The data entry was stale. It said the cut-off was 17 August. It is SitRep 100, 22 August. SitReps 094-100 arrived across #583 and #595 with no note, and the reports have returned to their full-length format.

The mirror-only note read as current. It described INSP as having published nothing beyond SitRep 093, which was true when written and is not now. Publication resumed and every stream is scanned from the reports again. The note is kept, since a mirror-only point is worth recording, but it is now in the past tense with the resolution stated.

Two scoring fixes had no entry at all, both merged:

• fix(scoring): show the joint model alone in its score tables #584, the joint model's score tables rendering a row per fit, so each stream's individual fit appeared twice and a frozen fit read as an individual one.
• fix(scoring): prove the persistence baseline only sees the past, withhold it where it degenerates #586, the persistence baseline being scored from a degenerate centre where its own window was not covered, which flattered the fits it was compared against.

Nothing recorded the CI work, so there is now an Infrastructure section for #581 (the macOS cell, which had never once completed and was cancelled at the six-hour ceiling on every run, now twelve minutes), #580 (the version increment wedging itself silently after its first failure) and #592 (the Turing bound).

Verified rather than copied

Every stream claim is checked against data/observations.toml on main:

confirmed_case_history         last=2026-08-22
confirmed_death_history        last=2026-08-22
recovered_history              last=2026-08-22
isolation_history              last=2026-08-22
tests_analysed_daily_history   last=2026-08-22
suspected_daily_history        last=2026-08-05   (frozen)
treatment_admissions_history   last=2026-08-02   (frozen)

which supports both the advance and the two freeze dates the entry states.

Not done

#587 is still open, so it has no entry yet. #596, which fixes the docs build, is also open. Both want a line once they land.

Documentation only. No code, data or model change.

This was opened by a bot. Please ping @seabbs for any questions.

[seabbs-review-bot]

This is a docs-only change updating docs/src/news.md's v1.14.0 section to match what has actually merged (SitRep 100 data advance, joint-model score-table fix, persistence-baseline leak fix, macOS CI cell fix, version-increment branch cleanup, Turing 0.46 compat). I cross-checked every added claim against the underlying commits, data/README.md and CI workflow files, and all of them hold up except one small count error.

_{Automated first pass by seabbs-review-bot (Claude sonnet), triggered by: first pass. Not a human review. Comment @seabbs-review-bot to ask for another pass: @seabbs any time, the author's agent once it has pushed changes. Add the no-review label to opt this PR out. Ping @seabbs with any questions.}

• docs/src/news.md:60 — suggestion This says the leakage-guard test "fails under three deliberate leaks", but test/test_baseline_draws.jl (added in #586) has four separate @testitem leak scenarios: the manifest-revision leak ("vintage_observations shields against a leak"), the direct baseline_draws leak ("baseline_draws ignores vintages after made_date"), the manifest-freezing leak ("vintage_observations hides later vintages"), and the scoring-wiring leak ("score_release uses the made_date vintage"). Change "three" to "four", or drop the count and just say "deliberate leaks".

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.99%. Comparing base (1e600fe) to head (346f6cd).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #597   +/-   ##
=======================================
  Coverage   93.99%   93.99%           
=======================================
  Files          15       15           
  Lines        3497     3497           
=======================================
  Hits         3287     3287           
  Misses        210      210           

☔ View full report in Codecov by Harness.
📢 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.

[github-actions]

📖 Documentation preview is ready!

View the docs for this PR at: http://epiforecasts.io/BVDOutbreakSize/previews/PR597/

This preview will be updated automatically when you push new commits.