Updated the visualization docs to mention that the trust region limit…

…s are plotted with the cross sections when a trust region was used.

docs/visualizations.rst

@@ -64,8 +64,9 @@ Gaussian Process
   The color of the cross section corresponds to the parameter that is varied in the cross section.
   This predicted landscape is generated by the model fit to the experiment by the Gaussian process.
   Be sure to check after an optimization run that all parameters contributed.
-  If one parameter produces a flat cross section, it is most likely it did not have any influence on the final cost.
-  You may want to remove it on the next optimization run. 
+  If one parameter produces a flat cross section, it is likely it did not have any influence on the final cost or its length scale was not fit well.
+  You may want to remove it on the next optimization run or set limits on its fitted length scale.
+  If a trust region was specified, then markers indicating the upper and lower limits of the trust region for each parameter will also be displayed in the plot.
 
 - **GP Learner: Log of length scales vs fit number.**
   The Gaussian process fits a correlation length to each of the parameters in the experiment.
@@ -93,6 +94,7 @@ Neural Net
   The main difference is that the neural net learner will actually produce multiple instances of this kind of plot.
   The first few show the results predicted by each of the independent neural nets.
   The last one shows the average of the costs predicted by each independent neural net, as well as dashed lines showing the maximum and minimum values predicted by any neural net for each point.
+  If a trust region was specified, then markers indicating the upper and lower limits of the trust region for each parameter will also be displayed in the plot.
 
 - **Neural Net Learner: Cost Surface.**
   This plot is only generated if the optimization had two parameters.