Design: search results use §5's row and empty state, and no-hits leaves §5d #3649

For: the design audit (#2678). The search page should look like every other list.

Done:

• [ ] The search page's results use row() from src/web/ui.ts, and its empty result uses §5's empty state, in place of .no-hits, which src/web/style/content.ts defines today.
• [ ] .no-hits's rules are gone and the name leaves the 'search results' bullet of docs/DESIGN.md §5d (the bullet's other classes stay). bun scripts/design-lint.ts passes.
• [ ] The pull carries before/after screenshots of a search with hits and one with none, in light and dark.
• [ ] A test finds no no-hits in the page, and finds the empty state's words on a search with no hits.

Files: src/web/style/content.ts, docs/DESIGN.md, src/web/pages/search.ts, a test. Risk: none beyond looks.

Parked by the crew: the check failed in the place it ran, not in the change, and failed the same way when it was run again (pull #3818 closed). Planner: split this smaller, or rewrite what done looks like; a person can also say what to try.

Steward (trevato): pull #3818 was parked by an infrastructure fault, not this work - the egress cap from #3719 stopped every check's package install (02:20-03:12Z). The forge is rolled back and the cap is fixed (c7ea1935). Reopened and re-checked; this strike does not count.

Parked by the crew: the check failed ("a pull whose review loop is spent is the planner's (#1655) > the planner is asked for it and its brief says three reviews asked for changes") and the decider saw no point in another try (pull #3818 closed). Planner: split this smaller, or rewrite what done looks like; a person can also say what to try.

Parked by the crew: the check failed ("what you asked for > the page leads each row with what became of it, so a narrow screen cuts where and who, not that") and the decider saw no point in another try (pull #4017 closed).

Parked twice since a person last spoke here, so it is the planner's now. Every reason:

1. the check failed ("a pull whose review loop is spent is the planner's (#1655) > the planner is asked for it and its brief says three reviews asked for changes") and the decider saw no point in another try (pull #3818 closed)
2. the check failed ("what you asked for > the page leads each row with what became of it, so a narrow screen cuts where and who, not that") and the decider saw no point in another try (pull #4017 closed)

The planner splits it, replaces it with a fresh issue, or defers it; a person can also say what to try. Nothing starts here until then, and the board lists it.

Replaced by #4081: the same change, with the row order the 'what you asked for' test holds (outcome first, then where and who) written into what done looks like.