I agree, A way to disable missiles and reactivate them later would be nice
You can do that. It doesn't do it automatically, and the AI doesn't do it, but you can just switch your missiles to "no orders" instead of homing and they will ignore any intercepting missiles. Then switch back to homing so they engage the targets.
(Obviously this can be countered by sending a stream of missiles at your incoming wave, but it is a way of disabling your missiles. You can also give them a move order if you need them to jink to avoid an incoming wave of fire or missiles.)
The ship life question is actually interesting because it's built in to some of the reactor, engine, and crew module design parameters. The current assumption is that a minimum of six months is needed; no more, no less. It might actually be better if ship "shelf life" were variable, and different missions required ships with different shelf lives. Designing orbital gunboats that don't need to go anywhere might end up with very different constraints than something that might need to be on station away from support for years at a time.
It is deadly accurate at long range, sniping off weapon modules and radiators from over a hundred kilometers away on fleet carriers or gunships.
There are two problems here. One, that gun is bugged, taking advantage of an integrator error to create a weapon that is 3000% efficient. Two, a hundred kilometers isn't long range enough against heavy laser ships. I know johen's current lasership has something like nine 1 GW lasers (for less than the cost of a gunship, IIRC) that will fairly easily snipe that coilgun from 800 kilometers or more away. 2.50 centimeters of reinforced carbon-carbon is probably not going to work.
So far, I've been using diamond radiators simply because I haven't made a good study of it. Is there a good summary of what the tradeoffs are for using different radiators and radiator armor? Also, thus far my reactors have been the high temperature variants simply because I'm counting on lasers, countermissiles, and drones to destroy incoming missiles, and using my own lasers to defeat enemy lasers (so radiator laser vulnerability isn't really an issue).
It isn't a great weapon for every engagement, but it's adequate for a lot of them. Looking into some other ship designs that utilize extremely long-range sluggers or missiles combined with long-range point-defense guns like these.
One stress test to consider putting it through: see how it handles anything like its weight/cost in well-optimized long range lasers. Large laser arrays excel at disabling weapons at long range, and that single turret with only 3.1mm of boron carbide doesn't look like it will last more than a second or two; not long enough to get into engagement range even on a very high velocity intercept.
(Lasers have their own weaknesses, and can be swamped by large numbers of missiles with good anti-laser armor. Still, that kind of weakness would make me very hesitant to field this craft.)
Does the shape even matter all that much? At the velocities we're talking about, with the kinds of armor that are feasible to put on a ship and still have it be able to get anywhere, I'm not at all sure that kinetic munition shape is all that relevant. If the KKV hits going 10 km/s, the target is going to have a No Good Very Bad Day, whether the munition is shaped like a spike or a pancake.
I would advise caution rather than optimism. But it isn't helpful to simply dismiss the EM drive as impossible because it doesn't conform to your understanding of physics or assumptions about the drive.
It's possible that something genuinely interesting is going on, but the theories proposed by the EM Drive's inventors are somewhere between crank and con-man, and even assuming there's some genuinely physics-interesting stuff going on (rather than an experimental setup that's highly prone to different types of instrument error), that doesn't mean the EM Drive itself will actually be useful. In fact, given the implications of the claims, some extremely strong evidence needs to be presented to raise it above "this almost certainly doesn't work", and the evidence has been very weak at best. It's perfectly helpful to dismiss the core EM Drive claims because they don't conform to physics as we understand them (and we understand physics pretty good), for the exact same reason we (correctly) dismiss perpetual motion machines.
In practice, what would that bring? Antimatter thrusters?
As an RPG in a mostly-hard science fiction universe, Eclipse Phase doesn't concern itself with the details of space combat. Player characters are assumed to be more concerned with making a stealthy approach to insert teammates onto a space habitat rather than mixing it up with enemy gun platforms. The game describes space combat thusly: "There is no such thing as space combat: there is dying in space or not dying in space".
That said, Eclipse Phase brings a couple of different things, some easier to simulate in Children of a Dead Earth, some less so.
Eclipse Phase's main conventional thrust is metallic hydrogen. While containing the stuff is incredibly hazardous, MH has an energy density of something like thirty five times TNT, and a very high exhaust velocity. It has a specific impulse of 17 km/s, and craft with a MH engine are capable of SSTO on Earth (not that one wants to land on Earth, in CoaDE or Eclipse Phase, for good reason). It doesn't seem to use NTR rockets as much, as far as I can tell, but rather makes use of lower thrust fusion rockets for interplanetary maneuvers. Fusion rockets are described as having an acceleration of 0.05g, usually, and even conservative estimates of their exhaust velocity give them pretty impressive d/V. And, of course, military spacecraft often include antimatter rockets.
It uses a lot of speculative materials. Metamaterials can be used in the game for cloaks, but for space combat purposes they're more likely to be used for creating laser optics. Eclipse Phase also makes use of room temperature superconductors and very efficient high density power supplies. It's not clear to me at a glance how much this differs from some of the "stock" materials of CoaDE - the Eclipse Phase "materials" page mentions aerogels and metallic glasses, both of which are present in CoaDE, but as Eclipse Phase possesses full blown molecular manufacturing, it may have more optimistic or speculative materials available - I'd have to check with EP's creators and some other fans to be sure. Some of these properties are as easy as adding new custom materials, others enable revolutions in design that don't fit in with CoaDE's module maker.
The weapon differences include the superconducting materials available for electromagnetic weaponry, metamaterials available for conventional laser armament, particle beam weaponry, exotic lasers (gamma ray and x-ray lasers), micro/mini missiles (micro scramjets are a common infantry weapon), and antimatter bombs. Some of these are easy, but others would be very hard to include in CoaDE as-is.
Another large capability differences - widespread molecular manufacturing (along with smart materials and nanorobotics) makes it possible for Eclipse Phase spacecraft to build their own armaments and even crew.
Lore-wise, Eclipse Phase is a fully realized setting in a post-apocalyptic, transhuman space. There's a lot going on, and potentially a lot to base a campaign around. Space is intensely populated; every planet and moon, and many major asteroids have something going on, and all of the spaces in between are filled with habitats and spacecraft. In addition to planetary habitats, there are large torus and cylinder habitats with millions of people, ruined habitats and leftovers from the last war, cageworks for building spacecraft, migratory barges and cyclers filled with nomadic "scum", bioconservative armadas around Jupiter (and renegade bioconservatives with parts of that fleet), a cold war simmering between the autonomist belt and corporate inner system, fragile alliances within each 'block'...and extinction-causing posthuman artificial intelligence waiting in the background for the moment to act.