Fail the batch job when recording its end fails, and reconcile orphans

finishBatch now returns its error, and treats a batch that is no longer
running as one, so a batch job never ends succeeded with its batch still
running. A batch left pending or running with its job ended (process
stopped, or the final status write failed) is failed inside startBatch,
before the active batch is read, so it never blocks a manual or
replenish batch. This replaces the boot-only FailInterruptedBatches.

Document suggestions.name in the architecture schema summary.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

internal/suggest/suggest_test.go

@@ -891,23 +891,63 @@ func TestBatchFailsWithNoRegulars(t *testing.T) {
 	}
 }
 
-func TestFailInterruptedBatches(t *testing.T) {
+func TestStartBatchFailsOrphanedBatch(t *testing.T) {
 	e := newEnv(t)
-	queued, err := e.svc.StartBatch(t.Context(), suggest.TriggerManual) // its job is still queued
-	if err != nil {
-		t.Fatal(err)
-	}
 	now := store.Now().String()
 	orphan, err := db.New(e.st.Writer).CreateSuggestionBatch(t.Context(), db.CreateSuggestionBatchParams{ID: store.NewID(), Trigger: "manual", CreatedAt: now, UpdatedAt: now})
 	if err != nil {
 		t.Fatal(err)
 	}
-	n, err := e.svc.FailInterruptedBatches(t.Context())
+	b, err := e.svc.StartBatch(t.Context(), suggest.TriggerManual)
 	if err != nil {
 		t.Fatal(err)
 	}
-	if n != 1 || e.batchStatus(t, orphan.ID) != suggest.BatchFailed || e.batchStatus(t, queued.ID) != suggest.BatchPending {
-		t.Errorf("failed %d; orphan %s, queued %s; want only the orphan failed", n, e.batchStatus(t, orphan.ID), e.batchStatus(t, queued.ID))
+	if !b.Created || b.ID == orphan.ID {
+		t.Fatalf("batch = %+v, want a new one past the orphan", b)
+	}
+	if got := e.batchStatus(t, orphan.ID); got != suggest.BatchFailed {
+		t.Errorf("orphan status = %s, want failed", got)
+	}
+	// A batch whose job is still queued is not an orphan.
+	again, err := e.svc.StartBatch(t.Context(), suggest.TriggerManual)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if again.Created || again.ID != b.ID || e.batchStatus(t, b.ID) != suggest.BatchPending {
+		t.Errorf("second batch = %+v, status %s; want batch %s still pending", again, e.batchStatus(t, b.ID), b.ID)
+	}
+}
+
+func TestBatchJobFailsWhenRecordingItsEndFails(t *testing.T) {
+	e := newEnv(t)
+	e.start(t)
+	e.recipe(t, "Chili Verde")
+	e.reply(sideQuest("Chili Verde", "Your Beef Chili gone green.", "Beef Chili"))
+	if _, err := e.st.Writer.ExecContext(t.Context(), `CREATE TRIGGER refuse_done BEFORE UPDATE OF status ON suggestion_batches WHEN NEW.status = 'done' BEGIN SELECT RAISE(ABORT, 'done refused'); END`); err != nil {
+		t.Fatal(err)
+	}
+
+	b, err := e.svc.StartBatch(t.Context(), suggest.TriggerManual)
+	if err != nil {
+		t.Fatal(err)
+	}
+	job := e.waitJob(t, b.JobID)
+	if job.Status != jobs.StatusFailed {
+		t.Fatalf("job %s with its batch %s, want failed", job.Status, e.batchStatus(t, b.ID))
+	}
+
+	if _, err := e.st.Writer.ExecContext(t.Context(), `DROP TRIGGER refuse_done`); err != nil {
+		t.Fatal(err)
+	}
+	next, err := e.svc.StartBatch(t.Context(), suggest.TriggerManual)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !next.Created || next.ID == b.ID {
+		t.Errorf("next batch = %+v, want a new one", next)
+	}
+	if got := e.batchStatus(t, b.ID); got != suggest.BatchFailed {
+		t.Errorf("batch status = %s with its job failed, want failed", got)
 	}
 }