Status of the meta

Do you want module design tips or meta explanation? The things to min-max for are cost to effectiveness. How cheap can you make a useful anti-laser drone attacker missile, how cheap can you build an effective 1,000km laser (when including power and radiators), how cheap can you make a useful interceptor missile (must be cheaper than attacker missile). This forms the basic rock paper scissors of the current meta

Laser drones beat interceptor missiles (these are anti-missile missiles that are extremely cheap, lightly or not armored at all)
attacker missiles beat laser drones (these are more expensive than interceptor missiles, well armored against lasers especially from frontal attack)
interceptor missile beats attacker missile (they trade 1:1 but the interceptor is cheaper so you win by attrition)

Coilgun launchers are simply launch platforms for the missile itself (one can argue against/for their actual usefulness), NEFP warheads (or just flak / nukes) are just the payload for the missiles used, and railguns are the hipster thing as they require a metric fuck ton of micromanagement to do anything significant against a laser drone fleet since it will target one thing on its own and fire a few hundred useless rounds at it for 10 seconds while getting lasered to death. The only way to kill more than one target before being melted by a laser array is to manually micromanage the railguns spreading the damage around manually (probably one frame at a time with mass pausing). The problem is the AI will only switch targets once a target is destroyed, even assuming a really badass railgun firing 100km/s it still takes 10 seconds for the first projectile to land and attempt to kill the target (and likely will). Without micromanagement best case scenario for railguns is 10 second time to kill per target with no micro (and 99% wasted shots). With micromanagement it gets a little murkier. If railguns became more mainstream emergency high thrust boosters could be added to laser drones to counter them by doing a hard burn after melting the gun to attempt to dodge lethal (usually laser drones only have a valid lethal target zone of a few cm down the spine, you will likely lose a giant laser turret though, but 1 laser out of 20 for a ship is usually a fair trade (which is why I personally don't include them in the rock paper scissors atm).

Is cost currently based on anything except materials? It seems a bit off to me that a ship's armor will typically be more expensive than all the reactors, lasers, and other highly complex and expensive systems onboard. It doesn't seem to be a very meaningful measure of a ship's or module's cost if so.

It breaks down even further when you compare the price of modules to the price of fuel, but there's really no way to meaningfully and "realistically" represent manufacturing costs without a whole lot of assumptions and guesswork.

Which, in the setting of CDE, we probably are.

Design tips. What material properties do I want for a given purpose, do I use NTRs for missiles/drones or not, what would be a good way to test new ideas, etc.

One thing I've also started doing, when I use someone else's designs (I almost exclusively use apophys' reactors, for example), is to name the component after the original creator.  It keeps me from getting sued by Apophys Incorporated for copyright infringement.  Kidding aside, it's a way to honor the hard work of the min-max community.  

I would one day like a way to "subscribe" to other people's products (like steam integration) so when I toss on their coil gun it gets updated as they continue to update the design as well.

It's counterintuitive if you know what works where IRL, but in CoaDE fission tech scales downwards very well, so there really isn't anything too small for a NTR. NTR aren't litterally always the best thrusters for your drones and missiles, but they are the best thrusters in the game outside of very specialised cases (MPD for high endurance is the best known exception).