Allowing manual control of build queue defeats the whole point of sectors though.
I don't think it does, not conceptually. As they are now, sure, but if you're willing to do some substantial reworks... Personally I don't mind sectors currently - just stick a system in one and stop caring about it - but it's not interesting and I can absolutely see why people have problems with all the AI/building/failure to be trusted to utilize tile resources intelligently.
A better approach I think comes down to focusing on how sectors were introduced in the dev diary talking about them: as something similar to CK2 vassals.
TL,DR: Dropping the focus on sectors being a player aide and instead being primarily concerned with making them a political mechanic only makes for better sector mechanics overall.
CK2 vassals are
not primarily tools to assist with management from the player's benefit (although, yes, they have that as a bonus side effect - if you play with demesne limits turned off managing all your counties gets very dumb very fast). CK2 vassals represent the difficulties of managing a realm from an in-game perspective, force you to pick which areas you directly control and which areas you control indirectly (and thus receive less benefit from in terms of taxes/levies) and introduce internal politics that make the game more interesting.
CK2 vassals do
not restrict you from spending your own resources on developing vassal holdings, or on creating new barony-level holdings. Granted, that's far less involved than a tile-based system (and province development isn't the
point of CK2 in the way that pop/tile management can be said to be the point of Stellaris), but it's I think a good thing to note that the relatively recent ability to be able to develop any of your vassals' holdings was touted as a feature and was very well received. While I respect that there could be a vision for sectors being far more autonomous than CK2 vassals in terms of development (and having such a design vision and wanting to make that a reality does have some intrinsic merit), at some point that needs to be weighed against the practical concerns of what the players want. Giving players the option to influence what sectors build while still allowing sectors to do most of the work on their own if the player lets them is not a bad thing. I mean, you've already recognized that AI control of slavery hasn't worked out, and we already have the ability to move pops around in a sector-controlled planet for free - there's not a lot of justification left for not giving players the option to build their own buildings if they choose to.
So, with that covered, how should we approach sectors in Stellaris?
First, stop marketing them as player aides. As long as they are marketed that way, players will expect them to perform as good as a human player in order to feel like it's actually an aide and not a liability. Getting the AI to that point is almost assuredly not feasible - those who care will never be satisfied, and you will always see people complaining that sectors need to be "fixed" as long as they don't perform the way individual players think is "best." There is no amount of settings you can give to a sector that will make them appear to be competent to all players in all situations. Additionally, this isn't something everybody thinks is essential: some people like tedious micro (or at least don't have a problem with it). Not giving way on this is probably a bad hill to pick to die on. There's just not a lot of enthusiastic support for it - at most you've got tepid acceptance, and some real vitriol from those who are very unhappy with it.
Rather, the focus should be on the fact that sectors model the inability to directly control an entire empire (Yes, people who want to be able to take the entire stockpile of resources from a sector for free are objectively wrong. Move along). Sectors should also be made interesting politically. We don't know yet how the new faction system plays, so this isn't something that can be fully elaborated on yet, but there could be some options there. For now though, I'll assume the following suggestions aren't going to be related to factions directly.
Sectors should have governors assigned automatically and for free (no influence cost, and they don't take up a slot in the leader pool). However, governors should have some opinion of how things are going in the empire. I don't have concrete suggestions for what should affect opinion at the moment, but if we're just spitballing here we could have things like relative strength of sectors (a governor of some backwater sector won't be happy if you keep giving systems to an already rich sector), tax, winning/losing wars, personal ethics of the governor, adding/removing systems, etc. Unhappy governors will govern poorly, can foment unrest in the pops in the sector, work to decrease other governors' opinions of the empire, and if they are unhappy enough they can revolt. Influence can be spent to increase opinion, and influence can also be spent to replace existing governors or choose a possibly better governor than the auto-generated choice. Governors you personally appoint will have a higher opinion of you than the auto-generated choice. The automatic governor choice should have ethics respective to the pops in that sector, while ones you appoint have your empire's ethics. Pops appreciate a governor with their ethics but will be unhappy with a governor of different ethics, while a governor of the same ethics as your empire is easier to keep loyal. Replacing a governor before their term is up (either death or possibly an actual term, depending on the government of the actual empire or whatever decision you think is relevant) will also decrease the opinion of other governors (and replacing an unhappy governor makes it very likely for them to revolt instead of go quietly). Sector revolts should be dangerous, with the
old CK2 revolt mechanics of being able to call in other sectors after the revolt has started, and potentially other external empires should be able to join in and support independence of an ongoing revolt.
The player should
not be able to directly set the absolute tax rate. Rather, like in CK2, you can set the expected tax rate and opinion determines how much of that expected rate you'll actually get. You
can set the expected tax rate to 100%, but doing so costs a continuous small amount of influence. You can also take resources from the sector stockpile, but doing so should cost influence and impact governor opinion. The sector stockpile should also determine the size of navy that will be spawned if they revolt - an unhappy governor is likely to have a bigger stockpile to work with due to not paying tax, so they can actually become dangerous. The one reasonable complaint I've seen people give about not being able to use sector resources is that when the sector hits their resource cap all those resources become entirely wasted, with no option for the player to do anything about it as opposed to the empire resource stockpile, where you can always just spend it. Rather than letting the player use this directly, I think any surplus resources a sector earns that they can't store should go to slight increases to pop happiness - say it's resources that can now be spent on public works, or bread, food, and circuses or whatnot.
Finally, we need to address how the new sector mechanics would impact influence. Since a lot of these suggestions use influence, and since influence is still needed for a lot of non-sector-related uses, we might need to get a bit more influence income than the current static maximum. Since all these new influence costs are related to sectors, the best way to make influence income scale appropriately is to base new influence sources on sectors. Specifically, we make it governor opinion. A supportive and loyal governor (if we have an opinion scale of -100 to 100, say >= 50) will actively work to support your empire politically, giving you influence. Similarly, a severely unhappy (<= -50) governor will actively work to undermine you politically, costing influence.
The end result of all of this is that sectors are no longer simply player aides - and non-optimal aides that many feel are forced and unwanted, or at best just there. Rather, they're an interesting political mechanic that you work with because doing so is rewarding, not just to avoid micro. Sectors that you don't tax aren't simply wasted resources, they're now happier governors and therefore a more stable empire. Sectors don't feel like an arbitrary barrier to good planet development, but no matter how you develop you still can't make use of that full development for free - or at least that benefit is only localized to that sector in terms of happier pops. Making sectors is both something you want to do (for the potential influence) and potentially risky for internal politics. And as a bonus, they still will retain the micro-reduction benefits if you just let them be autonomous even though that's no longer the primary focus of them.
