That assumes that warp fields and the nature of subspace operate in a manner in accordance with Einstein's Theory of Special Relativity
No, it assumes that relativity of ANY kind is basically correct (Einsteinian, Lorentzian, Maxwellian, take your pick) and that the normal laws of motion are at least slightly applicable to starships. Special relativity doesn't apply to warp drive because warp fields represent a non-inertial reference frame and would therefore fall into General Relativity.
The prime issue pertaining to the use of Relativistic style effects and mechanisms is that Roddenberry and co. threw those particular features of realistic space travel out the window on Day 1.
Cool. But the basic concept of relativity still applies, and is the cornerstone of all modern physics including and not limited to Einstein's thoeries. That being, that all measurements are only said to be valid in the specific coordinate system of the observer. Lorentzian Relativity, as an example, postulates time as a fixed coordinate and instead reconciles observational differences through contraction of
distance rather than the dilation of time. Warp drive could very well obey a Lortentzian system, considering how often we have seen the Enterprise either accidentally fold, twist or even tear space with its warp drives.
Simply put, a starship still has to output power from its M/AM reactor to maintain whatever velocity attained
Measured relative to WHAT? The center of the galaxy? The planet they left? The luminiferous aether?
That's kind of how it's depicted because the writers Did Not Do The Research, but it makes a lot more sense if you simply retcon it as a measure of acceleration and the exact flight dynamics of warp travel are simply not fully or accurately described (which, either way, they're NOT).
This assumes that the system is generally Sol-like and while we can assume that the majority of systems are similar, that is not necessarily always the case. Given that, objects in the outer solar system (the Oort cloud) are fairly inconspicuous, and there are alot of them, colliding with one would generally be bad.
And yet the odds of accidentally running into an object in the kuiper belt or the oort cloud are
astronomically small, even at FTL velocity. We call it a "cloud" but the average distance between any two objects in the oort cloud is something like 200 AUs even even at its inner edges. The Kuiper Belt is even more diffuse; despite its far greater mass, its density is less than a thousandth of the asteroid belt, where accidental collisions are already incredibly unlikely.
Likewise, a single particle of gas and dust striking a ship's hull at c will definitely have a detrimental effect on its occupants and systems- hence the use of the Navigation Deflector System...
Which renders the ship totally immune to those kinds of problems.
As it stands, a single particle of gas moving at the speed of light (hydrogen or helium atoms, let's say) is not actually that big of an issue. Conventional space ships with no shielding get hit by those all the time in the form of alpha and beta radiation. Cosmic radiation is believed to be lighter particles at much higher energies. Traveling at very high FTL velocities would simply add energies to those particles and they would strike the hull as a constant rain of high-energy radiation, and a shielded starship is fully capable of just shrugging that off without a problem.
There are all kinds of flight sims and space sims where you can put some of these assumptions about speed and distance to the test. Celestia is a good one for this; pick a random direction and fly there, and you're unlikely to run into anything AT ALL, even if you have all the known asteroids and comets plotted. The same thing happens in Elite Dangerous, which does a pretty good idea of simulating the positions of asteroid belts and even the Jupiter Trojans in the Sol System; it turns out that, when cruising through an unfamiliar system, accidental encounters with planets are not only very rare, they're also totally avoidable with the smallest of course corrections, and are actually kind of difficult to pull off if you don't already have the planet/asteroid's orbital data in your computer.
Space is BIG. Planets, in the grand scheme of things, are tiny, tiny things.
