fix(server): remove stale and duplicate comments in semsearch queries

Remove a misleading "Simple example" comment on the production scoring
logic and two duplicate "Re arrange rows" comments that restated the
existing tie-break comment.

Author: peteski22

server/backend/src/cq_server/semsearch/queries.py

@@ -254,14 +254,12 @@ def combine(relevance, distance) -> float:
             float: Combined score computed as 0.8 * relevance + 0.2 * (normalized distance),
                 where normalized distance is `distance / total_distance` or 0 when `total_distance` is 0.
         """
-        # Simple example: partially weighted over normalized distance
         relevance_weight = 0.8
         distance_weight = 0.2
         normalized_distance = distance / total_distance if total_distance > 0 else 0
         result = relevance * relevance_weight + normalized_distance * distance_weight
         return result
 
-    # Re arrange rows according to distance
     scored = [
         (
             combine(
@@ -281,6 +279,5 @@ def combine(relevance, distance) -> float:
         for u, distance in units
     ]
     # Match RemoteStore tie-break: score desc, id desc on tie.
-    # Re arrange rows according to distance
     scored.sort(key=lambda item: (item[0], item[1]), reverse=True)
     return [u for _, _, u in scored[:limit]]