Wasm will likely be the browser target.
> 2. Any reason targeting QT isn't on the roadmap? [targets for KDE, Sailfish, ect..]
They wanted to go native to start. QT is GPL licensed. Not a good fit for Red. Their "small" installer, which downloads the big stuff, is 18M. Red, with self-contained toolchain and cross compiler + GUI, is 1MB.
> 3. Any reason wxWidgets wasn't a great choice as a foundation? (I believe they have a C API last I looked)
Targeted at C++, and also very large.
> 4. Is there or will there be an API for creating custom languages that can target/leverage Red's infrastructure? (basically a Red version/wrapper for flex/bison)
You can use libRed, which makes Red available as a DLL you can call into from other langs, like C. See http://www.red-lang.org/ where a couple GIFs show integration with VBA and Excel even.
Ultimately, the decision not to use something comes from value added and available resources. Team Red is small, and has to pick their battles carefully.