Its not a no-brainer to add this stuff. Every new display option we add gives us more code and options to have to support. more possible bugs and glitches.
Of course, every new feature brings new code and so the chance for new bugs and glitches. But is that a reason not to implement any new feature anymore? There’s a reason for new features. User demands.
You cannot fit 100% of the user base with a new feature. Some will use it, some will not. And initial useage is also no real proof if a feature is useful. Some features grows from no go to absolute favourite over the time. Ngons anybody? 
I meant the user base here with no brainer, not the coder base. The feature with the wireframe colours is a no brainer because it is useful for quite a few people, while it doesn’t harm the rest. You have already the code working as i can see. So even from here it’s a no brainer, you have already mastered the technical aspects. You just have to put it into the trunk now. Which is of course again work. But that one should be a smaller step now.
What’s left is the question, is coloured wireframe useful enough to put it into the trunk? From my angle this question is already answered by the often repeated demand, more than one person in this thread here (the image from NinthJake above says more than a thousand words), and even more by several implementations in other software.
As a contrary, tell me the usefulness of layers. They don’t get used by everybody neither. But are still part of Blender.
When layers are in Blender, then in my opinion colours for the layers / wireframe colours should be in Blender too. Because it’s a very useful sub feature for layers. To separate the content of the layers visibly and make the orientation faster and easier. To be able to tell, this object is in the green layer, and this object is in the red layer.
Now think of a motorbike, made of let’s say 20 layers. And you want to tweak at the motorblock mesh. Without wireframe colours you have everything in one colour. And because everything is one colour you cannot say where one object ends and where the next starts. And this becomes even harder the more dense the mesh is. Up to the point where you only see a opaque surface instead wire lines. How will you judge here if a edge is at the wrong place then? How judge if everything fits?
To select the object, in our case the motorblock, is again one of those nasty workarounds. It separates the selected object by a different wireframe colour (here you have your proof of being useful already, and it is even already implemented in Blender) . But that way you can only see the currently selected object visibly separated. With wireframe colours you can see all objects / layers visibly separated. A big relief to judge if a part doesn’t fit and needs more work. A big relief to grab that part then too.
If this is so obvious then someone should be able to explain why its useful.
Tough one. You already seem to have rejected the reason of visible separation and easier / faster orientation already, which is like “explain me why daylight is good” while you reject the reason “because then you can see” . That the same feature exists in other 3D software is also no reason for you as it seems. And also not the often repeated demand. There’s nothing left anymore.
