I made a 1 hour video of this, but I feel it is probably painful to watch lol, and you probably wouldn't want to, but if you want me to I can upload it.
My thoughts are as follows.
I can't help but feel the solution is overcomplicated. Switching between modes sounds straightforward enough on paper, but when it changes everything about the controls and behavior of the craft rather than just one or two things it makes things much more complicated. And for example, if I want to control vertical thrust, I need to change to hover mode, but in hover mode I can no longer roll the aircraft, and etc. By gaining some things I'm losing others.
Also, "automating" certain things to smooth out the user experience can be a great tool, but while the goal is to simplify the inputs, if you're not careful you can make them more complicated in a way. For example if I just want to move straight forward, I switch to manual mode, but if I don't have enough speed it's only giving me vertical thrust. I "can't" just move forward because I have to contend with the rules of how it wants to work rather than how I want it to work.
Between this and the aforementioned mode switching, I find myself wanting for more consistency in the controls. My game is designed as a very different experience, but it also features both airplane-like and helicopter-like control. I have no mode switching; if I want to move up I press the button to move up; if I want to move forward I press the button to move forward; some things adjust here and there but the inputs always move you in the same ways and when you want to move (or at least apply forces) in those ways you always can. It's hard for me to imagine why that wouldn't be achievable for what you're trying to do, but maybe there's something I'm missing.
The thing I had the hardest time with in general was slowing down. There are no brakes, and while I didn't specifically expect there to be brakes while playing, it does beg the question: how does a craft that flies like an airplane and lands like a helicopter transition between the two without brakes? It clearly is possible to do so within your game, but it's not clear what the proper answer to that question is or what the intended method is. Of course the other question you can ask is, why not have brakes?
Edit: I just found the part on the game page where you mention the engines firing backwards, but the way you implemented it seems counterintuitive. Why not give the player more direct control over it rather than have it tied to a particular maneuver in a particular case?
I wanted to do something like fly in and bank sharply left or right and then use vertical thrust to counteract my forward inertia, but switching to hover mode cancels out my ability to roll.
When I was in the tunnel I was trying to switch to manual mode and fly straight forward to exit the tunnel, but it kept flying me up into the ceiling of the thing, because I was not up to speed. It doesn't seem helpful for the manual mode thrust to move you up vertically when you're already up in the air but not up to speed. I also sometimes would try to accelerate and roll/turn right away but since the thrust starts out vertical only I would just go straight into the ground. As I understand it you're trying to compensate for a lack of lift in the transition to forward flight, which understandably complicates things, but it brings me to my next point.
While I'm confident "airplane controls" are a good idea for a spaceship game, I'm not so confident that it makes sense to have aerodynamic lift. I could understand if your goal was to create a game about a VTOL jet that can also somehow fly in space, and you wanted to achieve that with some realism, but the way you presented the concept was wingless sci-fi movie space ships taking off and landing, which doesn't seem in line with that. I understand there is a collision there between making something fly like a plane and not fly like a plane, but there are certain things about the concepts of gaining and maintaining lift that I don't feel someone piloting a spaceworthy metal brick should ever have to think about.
I definitely believe in the idea you have here, and clearly you've made it work for yourself, as the videos you've shared look very pleasing in how you're able to pull off the landings.
