But do LACs, even dedicated ones, have enough computing power to control hundreds of shipkiller missiles in a salvo and accurately target the enemy ships? They obviously can't be within 1 million km of the enemy, otherwise the enemy will simply blast them with energy mounts (or even counter missiles). A LAC that lights up active sensors will be targeted by the enemy at any distance, especially if it's helping the shipkillers. It may be too late for the first salvo, but any enemy survivors will have a lock on the LACs
If you're staying 'near' your own fleet, then there's a real question whether just using that volume to put fire control processing power on the ship, rather than putting it in an LAC, which sits in a bay, essentially giving you far less volume available for the volume invested.
Where an LAC - or drone, or whatever - near your own fleet is useful is as a
relay. Firing missiles is a pain in the neck because a wall of gravetic wedges can't be easily seen through and it's hard to talk to the salvo(s) in front of the last one launched, or clearly observe the target, as a result. The analogy of 'gunsmoke' was used in one book (I forget which but I'm sure someone will tell me).
One big part of ghost rider was the 'off axis' fire control relay, and friendly LACs might be useful in the role for a broadside-heavy fleet.
That's probably less of an issue when you're using pod-based launchers; if relatively few (if any) of your missiles are coming from your broadside, then your broadside is 'clear' and can see and communicate with the missiles pretty easily.
Where a 'fire control LAC' might be useful is in
forward fire control. Sending an LAC in close to the enemy fleet is....pretty suicidal (as you noted, going within energy range is guaranteed death), but if you don't have the equivalent of an Apollo control missile and FTL comms, then the only way to have meaningful accuracy with MDMs (or cataphracts) is to have someone closer to the other end of the missile flight with a fire control computer and a datalink.
Getting up there isn't actually that hard, provided you're not being over-optimistic. Getting to, say, standard missile range sounds doable. SLN stealth tech has been repeatedly mentioned as the one thing that isn't too far behind manticoran-level tech.
The big problem is that LACs without Grayson-style compensators and fission plants can't use heavy energy weapons to threaten capital ships, and won't have the acceleration to act as 'dogfighters' against enemy LACs, but if they were prepared to hang off at 'standard' missile range, it's possible that on silent running and using stuff like whisker lasers you could provide last-minute fire control that would seriously improve the terminal accuracy of a cataphract.
Obviously if they get spotted, they're dead, and relying on them being there is therefore a bad idea, but it feels like the most useful thing you can do with an LAC.
The question is if you can provide enough sensor detail from passive sensors to meaningfully 'help', what range you can stay covert at, and what bandwidth you can provide a datalink to missiles at without advertising your position to the enemy.
Obviously, the other big use for LACs is countermissile platforms.
They're cheap, expendable, and put a lot of CM tubes into a region of space that don't have to worry about interfering with broadside salvos. Given that SLN designs are supposed to be catastrophically light on missile defences, converting a single ship in the formation to be a carrier with CM-armed LACs is probably a lot easier than refitting every SD with additional CM tubes, or giving up shipkiller tubes to throw CM canisters.
(Yes, I know they're still going to lose against modern designs)