A standard ship turret isn't massively bigger than the tank turret shown. And, for that matter, I've seen naval weapons stations with both missile and gun mounts - usually point defence type things rather than offensive weapons, but then the usual combined turret you see in traveller is a beam laser for use against missiles and a sandcaster for use against laser fire*.
By comparison, no, I can't see any reason to mix attack weapons in a turret because there's no real variation in target types - putting a particle beam shot through something works perfectly well whether it's a fighter or a light capital ship, and anything capable of long-ranged shots in the first place can engage either just as easily.
* Yes, I know. I'm not a fan of sandcasters because counter-firing point defence against speed of light weapons makes little sense. Don't Poke The Plot Holes....
Not sure why chin mounted would be better. A ship can point any direction in space while "flying" as they aren't using thrust based drives (at least post TL 9).
You can, and no, your facing isn't limited by your drives. It's more a case of giving the turret the widest sweep possible without being occluded by the hull - you can roll the hull to unmask turrets but that must take some increment of time and whenever you're doing that you're not pulling evasive manouvres as efficiently as you could be if you weren't.
Equally, theoretically your turret placement should be mostly concerned with allowing you to present your thickest armour and/or smallest target profile to the enemy. Of course Traveller doesn't have varying armour values and doesn't consider the size of a target in difficulty (given the orders of magnitude in hitting anything, I understand, but it's kind of random that I can hit a 10 dTon light fighter as easily as a 40,000 dTon heavy cruiser but a 2.5 dTon torpedo is apparently an impossible target...).
If you are launching a missile against a highly maneuverable target 10,000 km away, if it isn't smart enough to come out of a VLS type of set up, it has no chance of hitting. PURELY axiomatic.
I would both agree and disagree with that statement. If you can't
build a missile smart enough to come out of a VLS launcher, you can't
build one smart enough to hit a target at orbital range. I agree. Completely and unarguably.
But that doesn't necessarily mean you will or must build that capability into every missile design. Building a missile without a 'launch orientation stage' - be that stage a physically seperate drive or a slow-manouvring capability on the main drive - is simpler because you're reducing the complexity of the task the weapon needs to accomplish to "kick off the rail at 10G until you reach the designated 'target volume', then go active".
For military hardware where performance is king, I'm not going to argue, but a 'civilian' missile needs to get churned out for 750 credits a round less margin (in a currency not massively removed in value from dollars or pounds sterling, given the price of everyday items), which is a damn sight cheaper than even contemporary AAMs, whilst a launcher costs 1,000 times that price implies significant effort in putting the complexity into the launcher rather than the round.
One other though:, a concept seen more than once in sci-fi I've read is missile launchers which are themselves a light mass driver; the initial velocity of the missile is not due purely to it's own drive.
Admittedly, 6 minutes at 10G puts a missile at 35 km/s, so you'd need a serious hump for any significant portion of that to be provided by a launcher, but in a universe where ship-to-ship railguns are a thing, that is clearly possible.
