fix(monitor): serialize DB writes through a single drained writer

Every check spawned `go e.db.Save*(...)` with the error discarded: a
fire-and-forget goroutine per log line, check, state change, and alert
health update. SaveLog ran a full-table prune DELETE on every insert and
SaveCheck a COUNT + conditional prune on every check, so the hot path
amplified each write into several statements. Nothing tracked these
goroutines, so at shutdown they raced the store's Close() — writes to a
closing DB, silently swallowed.

Introduce a single writer goroutine that drains a buffered channel of
typed dbWrite values (log/check/state-change/alert-health). Writes are
enqueued non-blocking; a saturated queue drops and notes it in the
in-memory log rather than blocking the check loop. Write errors are now
logged instead of discarded. Retention moves off the hot path: SaveLog
and SaveCheck become plain INSERTs, and PruneLogs/PruneCheckHistory/
PruneStateChanges run on a 10-minute timer inside the writer (single
keep-newest-N-per-site pass via a window function). state_changes was
previously never pruned — now bounded.

Add Engine.Stop(): cancels the engine's context, then waits for the
writer to drain every buffered write before returning. main wires it in
before the deferred store Close() so no write races a closed DB.

SQLite gains busy_timeout=5000 and synchronous=NORMAL, applied via the
DSN so every pooled connection inherits them (a post-open PRAGMA only
touches one connection); WAL moves to the DSN too. :memory: test DBs are
left as-is.

Tests: writer drains on Stop, Stop is idempotent, and the prune queries
keep newest-N per site / N logs on real SQLite. Full suite green under
-race.

internal/monitor/dbwriter.go

@@ -0,0 +1,46 @@
+package monitor
+
+import (
+	"gitea.lerkolabs.com/lerkolabs/uptop/internal/models"
+	"gitea.lerkolabs.com/lerkolabs/uptop/internal/store"
+)
+
+// dbWrite is a single unit of deferred persistence. The engine enqueues these
+// onto a buffered channel; a single writer goroutine drains and executes them,
+// serializing all writes through one connection and surfacing errors instead of
+// discarding them. desc names the write for diagnostics on drop/failure.
+type dbWrite interface {
+	exec(s store.Store) error
+	desc() string
+}
+
+type writeLog struct{ message string }
+
+func (w writeLog) exec(s store.Store) error { return s.SaveLog(w.message) }
+func (w writeLog) desc() string             { return "log" }
+
+type writeCheck struct {
+	siteID    int
+	latencyNs int64
+	isUp      bool
+}
+
+func (w writeCheck) exec(s store.Store) error { return s.SaveCheck(w.siteID, w.latencyNs, w.isUp) }
+func (w writeCheck) desc() string             { return "check" }
+
+type writeStateChange struct {
+	siteID     int
+	fromStatus string
+	toStatus   string
+	reason     string
+}
+
+func (w writeStateChange) exec(s store.Store) error {
+	return s.SaveStateChange(w.siteID, w.fromStatus, w.toStatus, w.reason)
+}
+func (w writeStateChange) desc() string { return "state-change" }
+
+type writeAlertHealth struct{ rec models.AlertHealthRecord }
+
+func (w writeAlertHealth) exec(s store.Store) error { return s.SaveAlertHealth(w.rec) }
+func (w writeAlertHealth) desc() string             { return "alert-health" }