Construct 2 itself is a third-party solution to programming games.
If you want to avoid those, you should code it yourself in native language.
You are misreading what I meant: I really like Construct 2, and come from a developer's background myself. The thing that causes alarm bells to start ringing in my head is Ashley's response that it's up to Ludei to fix things. Who's to say Ludei will not respond in like manner, and tell us that Scirra should adjust their code?
Ludei's Cocoonjs is closed source as well, and we have to trust them not to change the license the day after tomorrow. Updates have been quite slow as well compared to other mobile game ecosystems.
For me personally at least, it creates a somewhat fragile environment to rely on for my development. That is why I have been investigating Ejecta as an alternative, and I got simpler projects running quite nicely through xcode. It also frees me from being dependent on Ludei, and I have far more control over the output. That, and the benefit of working with an open source, means I can fix things myself if required.
With Ludei I cannot fix anything myself, nor can anyone from the community. I do not have the time to debug to get our game working in Ejecta, so for this project this late in the dev process I just have to workaround known issues. Which I did today. It works quite well in Cocoonjs now.
For the next project I will absolutely NOT rely on Cocoonjs - that is what this first dev experience with Cocoonjs has taught me. Most probably Ejecta. If that proves to be unsatisfactory - well, lots of other options out there, including mobile game frameworks and libraries.
It will probably be years before out-of-the-box support for anything is added. Like I mentioned, making the game yourself in native language will provide all the solutions you're looking for.
Any ecosystem has its issues and bugs. I am very pragmatic about it: if one tool or ecosystem/framework has too many caveats, I will switch immediately. There's only one way to really find out whether it is robust enough or is lacking, and that's what I did: I relied for this project on Cocoonjs. Next time I will not! That simple. :-)
Cocoonjs tries to provide a non-technical way to publish to iOs, and I applaud that. It may work for others, but for me: I will avoid it next time.