Improve performance on Bragg and PDF's plotting

[wdzhou]

This PR implemented line color and style scheme for Bragg plots and PDF plots.

Fixes #75, #81,#85, #95, and #96.

== Test ==

1. Save S(Q)
   a) From tree item
   b) From combo box

2. Edit S(Q) by shifting and scaling from Edit S(Q) dialogue box.

3. Save G(R) by various file type.

[marshallmcdonnell]

The Bragg plotting is working great! Thanks for that.

There is one thing that I noticed that is not working in this branch that is still working in the "stable" version. When creating a new workspace via using the docked IPython widget, the workspace tree does not update with the new workspace. Thus, we can't plot up any workspaces we manipulate. The red box and arrows show the unsuccessful update:

Just to note, the same is true for the Bragg Peaks tab but this is an issue still in the "stable" version so it shouldn't hold up this merge.

[marshallmcdonnell]

You were right, the different alpha for same S(Q) but different configurations is a bad idea. Below is 4 different configs with the NOM_96771SQ_3 being an extreme case we would probably never use. Only then do we get enough change to see a difference:

Sorry but can we instead try the different line styles mixed with different markers idea? So maybe have the following dictionaries:

line_styles = { 0: '-', 1:'--', 2:':'}
marker_styles = { 0:'', 1:'.', 2:'o', 3:'D', 4:'x' }

and then do combinations when we append a new configuration:

plot(x,y,line_style[i]+marker_style[j])

Just an idea though.

[marshallmcdonnell]

Another one, when pressing the 'Save G(r)' button in the 'Calculate G(r) Tab', we get the following error:

[marshallmcdonnell]

Ready for merge.