So, A) I understand endgame. I've cleared turn 5 to bro. To be frank, Pet based abilities would probably be similar to what we see on SMN atm (Either buffs to make the pet hit for more, pet ultimate attack, or pet moves that modify your moves). These abilities would likely just fit in new places in the basic lancer rotaion and not themselves be positioned based, thus not really impacting the position play. Just sort of logistically, why would it matter which side of the mob I'm on when I give my pet a verbal command? Revisit this later. Also, as for micromanagement... I mean I've got schi88 and I've done all content on it as well. So.. yeah I feel you but I think you are overestimating the challenge.
B) Reducing base damage isn't a problem and doesn't pose a threat to current lnc/drg play. Its totally okay to make its so our WyvernGoon receives a sort of stance buff when it calls the pet. Ex: Summon Wyvern: Calls forth your wtfamazeballs friend, reduces personal damage X%. There is no problem here. When I stated that they had "solved" this issue for 2.0 I simply meant that Summoner on its own does less damage than a black mage, but when summoner uses its pet, together smn + garuda egi = (approximately) a blm's damage. Point here being: SE has clearly learned how to balance pet and master over all damage to be commiserate with other classes.
C) MRD and GLD dps are actually pretty close to each other (both classes have nearly the same possible enmity generation)... so yeah there's that. I'm full i90 both classes, tanked every fight in the game on both. That said, I've no idea how a pet-tank class would work, but could be cool.
D) I don't see how the story is a problem. YP and SE have said they want to expand the current armory system and deepen it, adding at least two jobs for every class, and even adding "advanced" jobs that need to be unlocked via quests or meeting multiple level requirements (such as both cnj and thm 30 for rdm). So sure, make the person have finished that part of the story line. Cool. Problem solved. Fluff is never an insurmountable barrier preventing content. That's the nifty thing about fiction writing, you just write more fiction to solve these problems.
E) The real issue isn't so much balancing the class or what have you, its having a reason to add it in the first place other than "we miss the pretty dragon give us fan service." Cus if all they did was reduce personal damage and make the wyvern make up that difference so you were just a different dragoon who got part of its damage from AI... there'd kind of be no point. The player would hit the same buttons, do the same things, just see wyvern animations along side them. It would be pretty trivial for SE to just code the wyvern to do X dps, and reduce potency on the weaponskills of the lancer by that same amount of an ideal or near ideal rotation. So.... lets talk about what sort of role it would need to fill, other than just staby staby.
E-1) At present, there isn't a very interesting buff/debuff meta game. Protect, adlo, stoneskin, damage over time.... sure some reduced healing received but that's almost entirely irrelevant in PVE. Lethargy and the other slow walk effects are nifty in turn 5 dreds.. but again pretty trivial elsewhere. If, however, WyvernGoon was made into a sort of mele-synergist/saboteur that perhaps used some sort of charge system (like warrior's wrath) to command the wyvern to use various breath magics, we could get a very interesting class. Completing the third step of a combo builds a charge, maybe 3 charges needed per WyvernSpellBreathShiny (or maybe only two, let SE balance that). Charge awarded regardless of positioning as long as the combo finsihes (of course, CT still has to be started from the back). This would heavily reduce the importance of moving for the WyvernGoon (Could just post up behind, sacrifice HT and build charges since they ahve reduced damage as is). Better players will have room to get more out of the class (continuing to move while managing the pet, but the focus of pet management requires only being behind)--and every class does and should have some room like this for better players to get more.
E-2) As to actual pet abilities, I mean... depends on what else SE is planning on doing. Maybe breath attacks enchant the target so that anyone striking the target receives the buff (Zephyr Breath: Players attacking the target receive a skill and spell speed buff for what ever amount and time is balanced, comparable to selene but buffs both, though obviously won't always affect healers). Or, could go with a battle pace theme, giving the Wyvern a slow debuff that can be applied (Not actually that unheard of since in 11 drg did end up getting power surge which granted a sort of haste) to enemies, with about the same total effect as ROH or Storms now. Maybe an HP or TP drain (TP would be really interesting). Could be a way to finally give us the en-fire we saw in the 1.0 trailer.
To be honest, it doesn't really matter what they ultimately add as the buff/debuff effects, just as long as they impact the group enough to keep the class viable. A spirit-link style move to maintain the pet and some basic placement commands will be enough. (perhaps something as simple as telling it to attack from front, attack from behind, attack from side, since it'll be mele like the wyvergoon rather than the full place that acn has).
TL; DR: Make WyvernGoon a support class, rather than a straight up DPS like Dragoon, and pretty much all problems solved. You can justify throttling the damage while still contributing to the group, makes it feel different from just dragoon, doesn't step on dragoon's toes, and adds diversity to our current play experience. Have the support moves tied to the pet and successful comboing and the increase in micromanagement relative to dragoon will be pretty small. (You'll have far fewer off gcd abilities and can spend that time ordering your pet).
I said add a new lancer job. Not sure where the confusion is. Let me be clear: This would be a second soul crystal equipped by the lancer class, just like SCH and SMN