With regards to this URL that you referenced above:
http://code.tutsplus.com/tutorials/using-createjs-preloadjs-soundjs-and-tweenjs--net-36292 This link looks like it is showing the output from Flash CS6 resultant from the publishing with CreateJS Tool kit.
On the other hand FlashCC has CreateJS "stuff" already baked into the IDE. All you have to do is set up a NEW FLA for "canvas" output. This output depends on the CreateJS tool kit when publishing. In addition, there are some differences between the tool kit for CS6 and canvas output in CC. In particular is the way the two handle imagery within the Flash environment. CreateJS tool kit for CS6 will bring each image separately. In contrast to FlashCC output, if you have two or more images, the code will automatically set up an image sprite sheet and an associated JSON file to tell the code where to "cut" programmatically to get the right imagery.
A sprite sheet for multiple images really helps in loading in assets; its quicker and it reducing the total number of server requests.
The preloadjs.min.js is simply code to aide in the loading in of the imagery assets. It will automatically be put into the HTML IF the code needs it.
I do this all the time with my creative work. I'll start out with crude PSD cuts of images that i need in my final out put. I will test publish to make sure it looks right. I will then, if feasible, make SVG (scale-able vector graphics) to replace the original PNGs, GIFs or JPGs I put into the FLA. After publishing to test, and if I was to remove all images for SVGs, my HTML output would no longer need that prelaodJS file any longer. And generally speaking if I made the correct choices (SVG for images or visa versa), my file overall size would also go down.
Hope this helps,
B