• Not sure it works but seeing that it doesn't have and shared I/O like the OG Fusion Reactor.
• Not sure on the fusion reactor thrusters since it keeps favoring the left side or is there a proper ratio?

It might exist simply because the reactor is copied from the vanilla one. I wasn't directly intending to include one but I'm not opposed to implementing it

it can be added since it can give a lot of options for layouts on oddly shaped ships especially of you are trying to make a very compact ship.

Currently it does show a neighborhood bonus, but the thruster doesn't actually care about the higher plasma temperature. Having 2 reactors per thruster looks strange, so I was hoping to affect the ratio with adjacency...

If the thruster accepted a plasma connection from the top (and either adjacency affected output rate, or thrusters cared about plasma temperature), it would even be possible to use 2x2 reactor arrays for the fusion thrusters...

Perhaps it would be a better idea to internally represent the thruster reactor as a nuclear plant (burns fusion fuel into heat) connected to a hidden "boiler" turning heat and water (/maybe another fluid?) into plasma production...

I think it would make a lot of sense for the fusion thruster to care about the higher plasma temperature, it would incentivize larger ships for more efficient thruster fueling.

this is offically planned once I get FluidEnergySource to work

I've been tinkering with this mod on my own and while I've learned quite a bit about the FluidEnergySource prototype, I don't believe there's a reasonable way to get it to work while also preserving the fluoroketone loop. It would feel pretty bad, but I believe the best(?) way to achieve the goal of having the fusion thrusters respect the fusion reactor neighbor bonus would be to just have the thruster itself take only water as input and instead of burning fusion plasma to convert into fluoroketone, have the internal assembler cost a large amount of power to run instead, and have that power either stay flat with increasing quality, or have that power usage decrease with increasing quality.

This is probably out-of-scope, but I wonder if you can use a "furnace" intermediary. Rather than hooking the Plasma output directly to the Thruster, hook it to a new custom Crafting Machine with several different recipes, each for a different temperature range. Each recipe would consume Plasma and produce as much Hot Fluoroketone as that plasma cost to generate, but would generate a different amount of proprietary "fusion thruster fuel" fluid based on fluid temperature.

Now this might be a bit clumsy to set up for the player if this were using the Assembler interface, but I wonder if a Furnace can be used for this. Furnaces pick their own recipe based on the ingredient, but I don't know if that process respects fluid temperature. Which is to say - whether it can choose between several recipes with the same fluid input but different temp requirements. That removes the need for hidden entities and means you can scale fuel production based on Plasma temperature.

The biggest bit of clunk here is that you end up with many duplicated recipes for the same thing, which... isn't exactly great practice. It also means that this can't scale infinitely, though I'm not sure how high up neighbour bonus can even go.

edit
Near as I can tell, you can't do much more than this with Fluid Temperature. The only things which respect temperature are generators (which can only generate electric power), reactors (which are heavily hard-coded) and ingredients/results, though there's no native way to link output to input temperature. In particular, Space Age implementations are pretty hard-coded, with very little customisation exposed for modding, unfortunately.

Maybe some composite entity where each part accepts a different temperature range as fuel, and each outputs "working fluid" to the same tank (which, if necessary, could have a limited output flow via a pump entity) could result in somewhat sensible behavior?

edit:
The "linking output to input temperature" doesn't seem to be hard (a single fixed-recipe assembler with "nothing + energy -> working fluid" and a fluid energy source set to scale_fluid_usage=true would produce it's output proportional to the temperature of the input fluid. The difficult part seems to be to guarantee that you get back as much fluoroketone as you turned into plasma. Since recipes have fixed product ratios, you can't increase working fluid per plasma without getting variable FK returns as well. Though I guess you could link an output pump from a hidden infinite FK tank to activate only while plasma was entering the machine (like a circuit flow meter using pumps and tanks). It wouldn't give FK back at the same time as the machine was consuming the original plasma, so eg. a blocked output would look the same as the machine getting starved of input in the tooltip (FK output full -> hidden plasma input pump inactive -> internal tank depletes until "out of fuel")