added note for the triangle inequality case in TSP (#6995)

Co-authored-by: Ross Barnowski <rossbar@berkeley.edu>

networkx/algorithms/approximation/traveling_salesman.py

@@ -224,6 +224,10 @@ def traveling_salesman_problem(G, weight="weight", nodes=None, cycle=True, metho
     the biggest weight edge is removed to make a Hamiltonian path.
     Then each edge on the new complete graph used for that analysis is
     replaced by the shortest_path between those nodes on the original graph.
+    If the input graph `G` includes edges with weights that do not adhere to
+    the triangle inequality, such as when `G` is not a complete graph (i.e
+    length of non-existent edges is infinity), then the returned path may
+    contain some repeating nodes (other than the starting node).
 
     Parameters
     ----------
@@ -265,7 +269,6 @@ def traveling_salesman_problem(G, weight="weight", nodes=None, cycle=True, metho
         List of nodes in `G` along a path with an approximation of the minimal
         path through `nodes`.
 
-
     Raises
     ------
     NetworkXError