Per-target score table

Maybe we should have a per-target score table somewhere. One column per target and then the rows are compilation time, run-time performance, native interop, debugging support and maybe more.

Like a feature matrix with checkmarks like (here)[https://haxe.org/foundation/support-plans.html] where 3 checkmarks = very good, no checkmarks = bad

We could take performance rates from the first graph at

So python would be in the bad group, then eval + PHP in the next one, then C# and HL and JVM and Node, then C++ and Java. Neko isn't working, it's probably in the bad group too.

Start by reviewing the proposed feature-matrix example and the formatter-noio benchmark graph linked in the issue. Define the target columns, score criteria, and data source before locating the website page to change; done means an agreed, maintainable per-target table covering the listed metrics.

Written by the indexing model from the issue text.