Do you want to have to rip out thousands upon thousands of thruster modules when you need to make a slight change in your thrust percentages?
It makes building a massive pain in the ass for no gain. It doesn't add anything to the game. It just makes it more complicated to build a ship in a game that's already got a massive learning curve.
Just the same as I have to rip out thousands of missile tubes if I want cannons instead? Yes. Does that make it complicated? I don't think so.
Wouldn't it take away something if you could just reconfigure your weapons on-the-fly to become cannons instead of missiles?
For me, having to intricately engineer spaceships adds value to the game, and conversely, not having to do so does take it away.
What you call a massive learning curve is part of what makes this game interesting for me. Learning and figuring out stuff is a considerable part of what I enjoy about the game.
Also part of my gripe with the current implementation is the user interface which I feel should not be necessary at all.
It might serve for an overall ship power configuration interface, where you could adjust power distribution to shields, or weapons, or engines in general, but I think propulsion is such an integral part of designing a ship that I should be required to put some thought into it before I build the ship, much less fly it.
And frankly, any argument about invalidating existing ships only goes so far before release. We've both seen balance and rebalance, addition of new and deletion of old systems (blocks as well as game mechanics) happen, and it will happen again and again until release, as it should.
We may not like some of them, but we'll start over and adapt.
While I can always draw some inspiration from my previous builds, I build in the knowledge that tomorrow's update may screw up all I build today.