All good points and cover many of the issues with early warp space flight.
Note that plasma cannot be corrosive as it is not acidic or alkaline. It is super hot so the term you wanted was errosive as it also has an abrasive property due to its excited state and forced flow.
Even in the age of steam, sail was the majority propulsion system despite being tempremental (doldrums), slower and dependant on travellingalong set routes (trade winds).
Errodable armour would be the priciplal defence against small object impacts at sublight speeds.
In the Rich Sternbach's "Space Flight Chronology" the UN commisons the solar fleet primarily to conduct search and rescue missions inside the solar system and maintain navigational equipment. Even during the start of warp powered exploration, naval vessels are still fusion and ion propelled with only a few scouts actually warp powered.
As far back as the DY-100 class fission propelled ships, Sternbach portrayed that the collison avoidance system was magnetic sweeps that removed dust and objects in the path by linear magnetic tractive effort.
Such a system would probally cope at 1/30th of light speed but not anything faster.
At warp speed any object entering the warp field has a synchronised velocity and direction with the ship producing the field. Its like throwing a brick into thje back of a passing pick up truck.... the brick adopts the speed of the truck and moves with it. The brick may slide about and contact other objects in the back of the pick up truck but its velocity, with regards the other objects in the truck will only be relative to each other and not in relation to the outside world from where they came.
If you threw a brick at a passing starship travelling at warp speed the brick would still retain the momentum energy you gave it when it entered the warp field but it would be isolated from the rest of the universe and move with the direction of the ship. (Golf ball rolling across the roof of a moving car).
If it misses the ship inside the warp field it will pass out the otherside , still carrying the velocity placed on it when it was thrown, but at a distance greater than could be achieved in the outside universe with the velocity it carries.
If an impact occurs between the ship and the brick inside the warp field, the impact velocity will be solely of the brick as the ship has a relative velocity of zero whilst projecting the warp field.
This was the problem caused by the asteroid caught in the Enterprise's warp field in ST The Movie. Inside the warp field the asteroid and the ship still had gravitational attraction with each other. The asteroid was moving at warp speed inside the field causing an overload of the warp drive as it struggled to project a field around both.
Obviously a decent navigational system or when you have one, someone taking notice of it, can prevent such occurances. This is probally why Star Fleet bans warp drive usage within the gravitational and orbital plain of any star system. Actually wasn't that Deckers's issue with Kirk's order to go to warp speed within the solar system??
The problem with converting non-warp ships to warp drive is on a par with trying to convert piston engined aircraft to jet propulsion. It'll work but not in a very satisfactory fashion.
There would be two distinct types of starships.... those designed for warp drive from scratch and those converted to warp drive. The latter would have poorer warp speed performance, power consumption and fuel efficiency.
Presumabily warp coils are constructed of some super conductor materials (as in they have an electron propogation velocity factor better than 0.999999999999999999999 with gold at 0.99, copper at 0.98, alluminium at 0.97 with 1 being light speed).
Obviously with Earth not yet venturing outside the solar system, such materials that are required to make the warp coils will be of limited supply.
Deep space exploration missions, sponsored by NASA's successor UESPA, would have taken years of preparation and expense. The loss of the SS Valiant (TOS Where No Man Has Gone Before) approximately 30 years before ENT would have been quite a blow and very discouraging. References to earlier Terran failures are made in ENT by the Vulcan delegation when they voice their objections to the NX programme.
Weaponry would be mainly missile dependant. If one's targets are, hopefully, as technologically limited and as slow as one's own warships and power is at a premium, missiles are a cheap and effective weapon system.
Direct fire weapons require power to operate. Targetting computers are required to calculate deflection angles, time dilation effects and fire the weapons (unlike the laughable fantasy of manually aimed Star Wars weaponry). Sensor and scanner systems capable of operating at relativistic speeds, including time dilation corrections, would also be more needed for direct fire weapons than missile system, which could carry their own independant systems.
Any direct fire weapons would be mainly for point defence against incoming enemy missiles.
RADAR and LIDAR could cope with the slower end of relativistic speeds but I think that even if Gateway's Super Amiga (NASA's computer system of choice) was the computer system installed, running the QNX Neutreno operating system, no computer would be able to cope with the reaction times needed as the speeds increased to half light speed. If Microsoft was involved you'd have to rely on crew members looking out the windows to spot enemy ships.
We could assume that between 2068 and 2120 some of this has been resolved, with some Vulcan help.
The placement of the warp reactor as close to or partially into the warp collar does explain the Vulcan starship design ethos in ENT. Note that the fuel and reactor are all at the bottom of the ships with the ring attachment at the same location. somebody's already reasoned this out when they designed the Vulcan ships for the show.
I would go for self contained fusion power source, warp reactor and warp coils all in a nachelle as a bolt on module, a bit like a big outboard motor that could be attached to anything, even a raft. It might not be very efficient but it does allow an instant warp drive conversion even to a DY-100. These could be mass produced, along with other essential sensor and scanner upgrades, as a kit of modularised parts. It would be the quickest way to convert a sublight fleet to rudimentary warp drive.
As for speeds attained on earlier starships, the Fortunate is quoted to reach warp 1.6 and the Horizon warp 1.8 in ENT. Both of these freighters is over 40 years old by 2156. If by 2156 the best speed is warp 5.2, a graph could be drawn from 2110 to 2156 to give an estimate of what, at best, could be expected.
Using the SFB / SFC Romulan impulse powered Warbird and Snipe as the benchmark of how many boxes of warp engine power would be represented by a fusion warp engine, in SFC terms, and guaging the relative hull sized proposed in a design in relation to either the Warbird or the Snipe, you have the movement cost scale.
I've finished my Donald Miller SSD inspired old frigate model so here it is. I've created five versions to cover its varied service life in the UEN Home Fleet, UEN Star Fleet, UFP Star Fleet and UFP Customs Patrol, which is where Donald Miller had it pegged for SFB.
I runs fine in SFC 1. Anybody wants to convert it to SFC 2 or SFC 3 be my guest.
Note that I've also put in all the hardpoints, something I sometimes have to do with other folk's SFC 1 models....