Ideally people would provide the PSD file (or whatever) which might include the UV map as a hidden layer, even if there are only 2 layers it's very handy. The other factor is for people learning, I mean just looking at other people UV mapping can really help when learning, see how the pro's layout their UV's and make use of the space available.
How do people do their texturing anyhoo? - like what is your workflow? - might be interesting to see how other people do this.
I use Lithunwrap to UV the model, then save it as a .3DS and save the UV map, usually as a 2048x2048 coloured bitmap. Lith can load most model formats, so people should consider it as it's free and can quickly provide a UV map image, even from a .X file. I load this UV map into PSP or Photoshop (I find PSP better for roughing out the texture), copy the image onto a new layer, get rid of the black background and make it a lighten layer. Then I can draw on the backdrop and create layers before the UV map, keeping the UV layer on top of everything. To preview the model I use Deep Exploration, which is damn useful, even the demo version because you can simply apply your PSD file to the model and see live updates - everytime you save the PSD file it updates in Deep Exp. You can also preview the model in different renderer's, like DX, OpenGL, and the jaggy software mode. I always keep the high res PSD file, usually just save as a .PNG then load it and rescale.
I never got on with 3D paint packages, I miss the traditional 2D art package layout too much.
We're going down... in a spiral to the ground...