Build the Review Queue chart's actual trace from per-second samples, not lap averages

A lap can span many minutes, so plotting one flat value per lap hid real
within-lap variation -- a single-lap hill repeat rendered as a dead-flat
line despite the pace/HR swinging throughout it. The actual pace/HR trace
now comes from activity_samples (per-second telemetry) when available,
looking up each sample's enclosing lap only for the target band/phase
color. Falls back to the previous lap-stepped rendering when an activity
has no samples.

backend/internal/api/review.go

@@ -22,6 +22,7 @@ func (s *Server) handleReviewQueue(w http.ResponseWriter, r *http.Request) {
 		store.KindAssignment
 		Activity store.Activity `json:"activity"`
 		Laps     []store.Lap    `json:"laps"`
+		Samples  []store.Sample `json:"samples"`
 	}
 	items := make([]item, 0, len(queue))
 	for _, a := range queue {
@@ -38,7 +39,15 @@ func (s *Server) handleReviewQueue(w http.ResponseWriter, r *http.Request) {
 			writeError(w, http.StatusInternalServerError, err.Error())
 			return
 		}
-		items = append(items, item{KindAssignment: a, Activity: activity, Laps: laps})
+		// Per-second telemetry, not just per-lap averages, so the chart can
+		// show real within-lap variation instead of one flat segment per lap
+		// (most activities only have a handful of laps).
+		samples, err := s.DB.SamplesForActivity(r.Context(), a.ActivityID)
+		if err != nil {
+			writeError(w, http.StatusInternalServerError, err.Error())
+			return
+		}
+		items = append(items, item{KindAssignment: a, Activity: activity, Laps: laps, Samples: samples})
 	}
 
 	// Most recent run first, by when the activity actually happened (not