P-M0-1: M0 changes no core-14 pass count

id:        P-M0-1
filed:     2026-07-17, migrated from the journal-format filing in the M0 milestone note;
           the settling run has not launched
suite:     core 14
tier:      free (deepseek-v4-flash-free)
engine:    tomo, tomo-cx, tomo-oi at the M0-landed pins, against the pre-M0 baseline
band:      floor: no core-14 pass count drops for any engine (baseline: tomo 14,
           cx 14, oi 13).
           ceiling: no pass count rises either.
mechanism: the M0 mechanisms were built on swebench-class failures; the core 14 were
           mostly passing already, so the honest expectation is a flat grade column
           with the wins showing up later on the harder suites.
settled by: the M0 exit smoke, core 14 sequentially per engine at the landed pins.
           go run ./cmd/lab run tomo && go run ./cmd/lab run tomo-cx && go run ./cmd/lab run tomo-oi
result:    (open, blocked on free-tier quota reset)
verdict:   (open)

This entry predates the ledger machinery. It was filed in the journal format inside the M0 milestone note, before the M0 gate run, and migrated here unchanged when the ledger landed; the gate run has still not launched, so the filed-before-run property holds either way.

A pass-count increase would be a pleasant miss and still gets a written verdict, because an unexplained improvement is as suspicious as an unexplained regression.