Problem: the digest script filtered by `number > last_number`, so any
upstream item with a number below the highest tracked entry was silently
skipped.
Solution: collect the set of all tracked numbers from `doc/upstream.md`
and filter by `number not in tracked` instead. For merged PRs, also
filter by the fork's creation date via `GITHUB_REPOSITORY` to avoid
pulling in ancient pre-fork history.
Problem: issues were split across four separate sections by status,
requiring moves between sections on every status change.
Solution: merge all issues into one `## Issues` table sorted by number
with an inline status column. Update digest script heading and row
format to match.
Problem: the upstream digest script referenced old section headings
(`## Open upstream PRs`, `## Upstream issues`) and a 3-column issue
row format that no longer exist after the tracker reorganization.
Solution: update headings to `## Upstream PRs` and `## Issues — open`,
and change issue row format to the new 2-column layout.
Problem: new upstream issues and PRs slip through because there's no
mechanism to surface them — manual polling of stevearc/oil.nvim is
required and easy to forget.
Solution: add a Monday 9am UTC scheduled workflow that reads the
highest stevearc/oil.nvim number from doc/upstream.md, fetches merged
PRs and new open issues/PRs above that threshold via the gh CLI, and
creates a structured digest issue in barrettruth/canola.nvim. No issue
is created when there's nothing new. Falls back to a 30-day window if
doc/upstream.md can't be parsed.
