ad 6) Pretender, the one-and-only pretender for Ashen Empire must surely be the Lich Queen.
While the Prince of Death is a nasty fellow who can fly inherently, the Lich Queen provides immortality (which even if she never dies is also recuperation and that's an awesome ability in its own right in a world that features eyeloss and armloss weapons amongst other things), a starting dominion of 4, good overall stats, and decently priced paths (sure, 40 isn't the best in the world, but is is better than many).
You can easily make her something like a mighty rainbow pretender (4 in most/all paths, higher death) with dominion 9 or 10, though less if you take high magic for better research. My personal choice of scales is the turmoil 3, sloth 3, cold 3*, death 3, luck 3, drain 2, going with the extra MR and points to spend on the pretender over better research (which delays the point where your research explodes which is when you start to forge skull mentors but certainly doesn't cripple you).
* exception: Cold 3 is usually the best climate choice but if you
know you are going to be facing tough players playing cold nations and less tough players playing hot nations you might reasonably choose Heat 3 instead. Every bit helps.
On awake/dormant/imprisoned...
The rational choice in almost cases would be awake. However, should you be about to embark on an MP game that, for whatever reason, you know or have a reasonable suspicion might be long (players involved, few players on large map, agreement beforehand, whatever), one interesting option is to go dormant, partially for the 150 extra design points, partially for the change it does to your opening gambit - the "low domain undead approach". (If it turns our you were wrong, and it is a short game, you are hosed - hey, it is a risky gambit with a potentially huge payoff)
With a dormant Ermorian pretender, your dominion will spread very slowly at start meaning fewer undead but also considerably more gold since your provinces will stay alive for much longer. With an awake pretender, you'll often be best off just pillaging provinces/taxing 200%, but with a dormant pretender it works really well to do a normal 100% tax, recruit every single mercenary company for insane prices (you can afford to offer more than anybody else as you have no real expenses), and just conquer as much land as possible with the undead you do get and the mercenaries, preferring farms over all others.
Depending on number of mercenaries available, you will be stuck at 1 province/turn longer than otherwise, which sucks, however, when your pretender does break free, you'll have 150+ death gems stored, and it will take you only 2-3 turns to get a research/recruitment factory up and running
1) Dusk Elder [total 1]
2) 2xDusk Elder [total 3]
3) 3 or 4 mix of Dusk Elder/Archbishops [total 6-7]
You'll hit the floor running with lots of provinces generating a few undead, your capital generating lots of undead, a huge income compared with what you'd have had with an awake pretender from turn 1 spreading your dominion like mad (not only are most provinces dying only slowly, they don't have your scales either), and not significantly behind in overall research since your mage recruitment earlier would have been limited by gem income anyway. This has allowed you, during the rounds until awakening and for the next many, many, rounds, to build way more fortresses than you'd have been able to afford with an awake pretender. As for temples, just start building them when provinces (with fortresses) reach dominion 9-10 to start raking in those Knights of the Unholy Sepulchre - no reason to build them earlier in this type of strategy.
An additional advantage is that with a domain that'll for the first large part of the game stay mostly within your own borders or even behind your borders you appear significantly less threatening to your neighbours - and to the graphs (if they are enabled).
The major downside, and there's no hiding it, is that you forfeit 10-13 rounds of sitesearching with your pretender (say one every 2 turns) - that's a really significant hit to your gem income that shouldn't be underestimated. On the other hand, it shouldn't be overestimated either, just like the ability to maintain a low dominion and high income for a long time shouldn't be underestimated where fortress construction is concerned - many undead from non-fortress provinces or fewer but from fortresses? You can call it either way.
Another downside would be if the game turns into a short game or a blitz, but then, the assumption of choosing this opening gambit in the first place is that it'll be a long game - that's just a risk to assess before play begins.
Ah, well, it has worked for me in one MP game and it might for you, but then again, it might not.
EDIT: It seems that some people have misunderstood the "low domain approach" to mean that the pretender should have a low dominion. Far from it! The pretender's dominion should be the usual 9-10 of Ermor (and more likely 10 than 9), the "low" comes into play because it takes longer time to spread it throughout your realm since it takes longer time to get going, spreading only via home province and temple check until your pretender breaks free. That is what allows you to rake in the cash to build castles with, castles that'll soon enough have maximum dominion generating lots of good undead. It is more money per turn for more turns with the expected long-term payoff of being able to afford more castles, temples, independent mages (if you find any),
while still generating the best undead troops in large numbers from fairly early in the game. I apologise for any confusion that may have arisen from my description of it as a "low domain approach".