Fix incorrect assumptions in Ruby li_std_set test

Examples/test-suite/ruby/li_std_set_runme.rb

@@ -57,7 +57,10 @@ s = LanguageSet.new
 s.insert([1,2])
 s.insert(1)
 s.insert("hello")
-s.to_a == [1,[1,2],'hello']  # sort order: s.sort {|a,b| a.hash <=> b.hash}
+#s.to_a == [1,[1,2],'hello']  # sort order: s.sort {|a,b| a.hash <=> b.hash}
+# Test above is flawed as LanguageSet sorts by each element's hash, so the order will change from one invocation to the next. Sort a conversion to array instead.
+sa = s.to_a.sort { |x, y| x.to_s <=> y.to_s }
+sa == [1,[1,2],'hello']
 
 EOF