implement 'npoints' strategy for the 'BalancingLearner'

Title is obvious.

adaptive/learner/skopt_learner.py

@@ -24,15 +24,17 @@ class SKOptLearner(Optimizer, BaseLearner):
 
     def __init__(self, function, **kwargs):
         self.function = function
+        self.pending_points = set()
         super().__init__(**kwargs)
 
-    def tell(self, x, y, fit=True):    
+    def tell(self, x, y, fit=True):
+        self.pending_points.dicard(x)
         super().tell([x], y, fit)
 
     def tell_pending(self, x):
         # 'skopt.Optimizer' takes care of points we
         # have not got results for.
-        pass
+        self.pending_points.add(x)
 
     def remove_unfinished(self):
         pass