Everything will get done, and not in the too distant future, but the order in which things get done depends on priorities the community sets. That's a constant feedback loop, driven by suggestions. That can evolve quickly as users start leveraging new features that emerge.
Radian came out because the big internal machine was immediately useful, as is, for data-centric people using SQL. It's natural that such early adopters would make suggestions directly in tune with what they were doing, thus driving the big expansion of dbms-oriented capabilities. As Manifold shifted gears more into the interactive GIS direction, and more interactive GIS features appeared, it made sense that more suggestions to expand the GIS end of it would start driving the process.
Earlier this year, I personally expected lots more interest in legends. I think many people did, hence the work on things like grouping logic that is a key element of many things, from grouped layers to good legends. But instead there's been much greater demand for filling holes in vector editing (more snaps, please) and expansion of raster analytics, with things like viewsheds. In response, those things I think bubbled up in priority ahead of legends.
There's a huge amount of infrastructure already built to support things like legends and expanded capabilities in layouts, with a lot of work being done on the big part of the iceberg that supports the visible tip (work being done to ensure sensible scaling of web servers in frames being an example). None of that is going away. It will all get done, and not in the too distant future. But the key thing to get layout features moved ahead of other things, like viewsheds, is to send in specific suggestions for what you prioritize. Be specific.
For example "features like legend, scale, north arrow, line and polygon graphics" is good, but I don't really understand what you mean by "line and polygon graphics" as features for layouts. I can guess, but maybe my guess is not what you want. So best to say explicitly.
It also doesn't help to say "like 8," because 8 has zillions of features that nobody uses, so the priority should go on doing better, 9 versions of those features that people actually use and must have.
If anybody is surprised at other priorities overtaking legends, an easy test is for everyone who is reading this thread to post a copy of the suggestion they sent in. One or two suggestions can certainly bump priority in a given direction, but not as much as dozens or hundreds of requests for something else.
Like I say, there's no doubt in my mind the spotlight will turn to layouts/legends/etc, either with a few features every now and then mixed in to builds that focus on filling in vector editing tools and expanding raster analytics, or as the focus of a few builds. Layouts in 9 are already more beautiful compared to 8, with way more elegant and accurate handling of things like rendering resolution. They obviously cry out for completion with just a few more features.
I'm just saying that specific "here, exactly, are my top five wishes for that, and here is how I want them to work..." suggestions are a big help in prioritizing those above other features that other folks want, to get them sooner rather than later.