There is some genuine thoughts/ideas in there. My recommendation start small.
There is a lot more at stake here than "lots of time" You are looking at a pretty massive tag list with very little consistency for shared assets. This creates an issue with tag space as multiplayer maps have a cap that is far less than sp so keep that in mind.
Additionally AI or enemy diversity does not equal a good game or play experience. I recommend reading this:
http://www.gamasutra.com/view/feature/131735/designing_enemies_with_distinct_.php It gives a basic overview and will help cull the list more. There are some good thoughts on what various enemy weaknesses or more roles are, but at the end of the day the document in it's current state is just a tag list. I applaud your desire to be comprehensive in approach, but Halo does not live on a google drive account, it live in the 30-40sec experience when playing maps.
In summation I recommend refining your list with a series of questions:
sp or mp?
if sp what kind of story and do I know how to script?
custom tags or classic?
if custom what do I want to use?
vehicle centric or weapon centric?
how large will my maps and encounters be?
*These are more quick start questions and in no way represent the cauldron of design questions any creator is faced with when designing or planning their games/maps. You are off to a good start in your approach, but really work on developing a focus with execution (30-40secs of gameplay) . In some cases this is simply knowing what questions you want to ask. :)