Track hist of environment reward

ml-agents/mlagents/trainers/stats.py

     std: float
     num: int
     sum: float
+    full_dist: np.array
-        return StatsSummary(0.0, 0.0, 0, 0.0, StatsAggregationMethod.AVERAGE)
+        return StatsSummary(
+            0.0, 0.0, 0, 0.0, np.zeros(1), StatsAggregationMethod.AVERAGE
+        )
 
     @property
     def aggregated_value(self):
             self.summary_writers[category].add_scalar(
                 f"{key}", value.aggregated_value, step
             )
+            if key == "Environment/Cumulative Reward":
+                self.summary_writers[category].add_histogram(
+                    f"{key}_hist", value.full_dist, step
+                )
             self.summary_writers[category].flush()
 
     def _maybe_create_summary_writer(self, category: str) -> None:
         if len(stat_values) == 0:
             return StatsSummary.empty()
 
+        if key == "Environment/Cumulative Reward":
+            full = np.array(stat_values)
+        else:
+            full = np.zeros(1)
+            full_dist=full,
             aggregation_method=StatsReporter.stats_aggregation[self.category][key],
         )