dupe ledger: #10208's suite_ledger_steward copies four bodies -- Duplicate Body Ratchet red on master #10216

Goal

Turn the Duplicate Body Ratchet (duplicate-bodies, tools/dupe_scan.py) green again on master and on every open PR. It has been red on master since 27efeba (#10208): "a body written under effect_wanted / run_starts_job / step_ef / writes_the_repo is now copied [2] time(s) and was in no ledger". #10208 added specs/ci/suite_ledger_steward.t27 and specs/jobs/suite_ledger_steward.t27, and they repeat four bodies from specs/tri/t27b/ledger_steward.t27 and specs/jobs/t27b_ledger_steward.t27. Seen on #10211 (run 39097511524), epic #10024.

Boundary

• Only tools/duplicate_bodies_baseline.txt changes. It is tool-written, by python3 tools/dupe_scan.py --bless, run on the Railway lab at master d9e0c6c.
• No spec changes. Replacing the copies with a use of the t27b steward's functions is the better long-term fix, and it belongs to spec(ci): suite ledger steward -- what the corpus ledger bot applies and refuses (Closes #10204) #10208's owner; this issue only records the copy so that every other PR is no longer red.

Scenarios

• Given master d9e0c6c plus the change, When python3 tools/dupe_scan.py runs, Then it exits 0 with "ok: no new duplicate body, and none of the known groups grew."
• Given the diff, When it is read, Then it adds exactly the four lines effect_wanted 2, run_starts_job 2, step_ef 2, writes_the_repo 2 and removes none.

Requirements

• FR-001: The ledger MUST be written by tools/dupe_scan.py --bless, not by hand.
• FR-002: The PR MUST change exactly one file, with 4 lines added and 0 removed.
• FR-003: No existing group's count MUST change.

Success Criteria

• git diff --numstat origin/master prints 4 0 tools/duplicate_bodies_baseline.txt.
• duplicate-bodies concludes success on the PR.

Close criteria

• The PR merges and duplicate-bodies is green on it.