This could become a religious war - there are strong proponents of Arducopter/Mission Planner and equally strong advocates for PX4/Q GroundControl. Personally, from an end user's point of view I think both systems show their roots in engineers/hobbyists and are not particularly friendly to use. From a flight point of view, a lot of the differences in opinion come from wildly different use cases for individual pilots. Simple things like whether you're comfortable switching between Mode 1 and Mode 2 can become major blockers in some people's eyes. Certainly a lot of people seem to fixate on individual features (never the same individual features!) when they complain about or praise a particular drone.
As it is, the decision to go with one or the other will have been as much a commercial as a technical one. I know Yuneec discussed the flight system with a number of external teams and chose the one they felt would deliver the best product, within the time and budget limitations they had. Whilst the H480 flight stack grew to be a reliable system, I understand that it runs pretty much at its limits and is very difficult to change to add new features. Yuneec's decision was about moving to a system that could support new developments going forward.
From my perspective, gliders' list:
1. Redundant compass/GPS - I'd like to see the failure rate before thinking this was a major issue. It adds a lot of cost and complexity to a drone.
2. Battery monitoring - Fairly irrelevant to me, I fly well within my battery limits and check the batteries properly on the ground with a balance charger. 'Smart' batteries don't help here.
3. Better flyaway protection - Again, I'd like to see the failure rate, but I believe PX4 has EKF filtering, and certainly innovation is to be encouraged
4. Fully autonomous flight - Given the current regulatory environment, no company can offer this in a consumer (or even semi-commercial) drone - it would be suicide.
5. Community based feature/bug tracking - This highlights the difference between AC (community based) and PX4 (commercial based) development. Ultimately though, the community support around a drone is down to the company selling it. Yuneec could not have launched a commercial product and told pilots to go hang out in the AC forums if they need a fix!
6. More RTL options - Certainly, so long as they have a good user interface to explain what's happening. As it is, most users don't understand their options or what happens in these cases and only find out when they actually have an emergency. Adding complexity can be a bad idea in a consumer product.
In short - yes, of course you can 'roll your own' and get a drone that does exactly what you want. You can't expect a company that saves you the time and effort that rolling your own involves to also provide a completely unique-to-you configuration. Yuneec add value by solving those problems. If they don't add value - by just sticking freely available components together - they'll make no money. We're seeing the same in the FPV market, where the free availability of flight stacks causes great innovation in the short term, but razor thin margins for the companies involved which ultimately stifles serious investment.
@PatR I completely agree with your points - it's down to Yuneec now to communicate what has changed in the company and what their goals are with the new machines. I don't like having to 'read the tea leaves' to try and guess what they are doing.