I'd add that while it's technically possible, for us as a small company I don't think we really have the resources to pull off official console ports. A very significant reason Construct has as many features as it does is because we truly have a single codebase that works on all platforms. There are also significant hurdles to writing native ports - this blog post on the prospect of native engines is quite old and a bit out of date now, but still covers some relevant points.
Even if we only aimed for a single console, it would probably take around 12 months, tie up basically all our development resources essentially putting everything else on hold; being web-first, many features would probably be infeasible to port (e.g. do consoles do iframes? SVG? form controls? video? networking?), making for a complex support matrix and increasing the difficulty of actually porting your games to console; and it would likely tie up significant resources even beyond completion, as console SDKs are a moving target with regular changes, and opening a whole new class of bugs where there are differences between the web and native engines. Meanwhile there are considerable risks: I'm still surprised at how few people appear to use or even talk about the Xbox One exporter, so it seems entirely possible we do all this work and hardly anyone uses it; maybe it takes so long that by the time we finish the next console generation is out and we have to start over; maybe they actually do add support for HTML5 games at some point making much of our work redundant; and due to all the risks and complexities involved it could end up being extremely expensive - if you think $99/year is expensive for Construct 3, note GameMaker charge $799 per year per console exporter, or $1500 per year for multiple console exporters - and they're a bigger company with more resources to put towards that.
Don't get me wrong, I'd love to have more console support for Construct. It would be huge and a dream for many people. However looking at all of the factors involved, it seems awfully close to an extremely risky bet-the-company gamble, and I don't think we can justify it. The de-facto setup of a couple of third-party porting companies who maintain their own engines based on compatible subsets of C3 is kind of awkward, but seems like the best compromise we have right now.