I can't be an expert in every GC implementation because there are so many of them, but many of the problems they mention are problems in those languages too. Finalizers are highly desirable to both the authors of the runtimes and the users of the languages, but are generally fundamentally flawed in GC'd languages, to the point that the advice in those languages is to stay away from them unless you really know what you are doing, and then, to stay away from them even so if you have any choice whatsoever... and that's the "solution" to these problems most languages end up going with.
Which does at least generally work. It's pretty rare to be bitten by these problems, like, less-than-once-per-career levels of rare (if you honor the advice above)... but certainly not unheard of, definitely not a zero base rate.
Latest version of C# is a fantastic choice for this. Java too but I would lean more C# due to the new delegate function pointers for source-generated p/invoke. Thing of beauty.
I also want to call out CppAst [0] and CppAst.CodeGen [1] projects. These two things have saved me years of my life if I were to roll these by hand. Kudos Alexandre Mutel, kudos.
Which does at least generally work. It's pretty rare to be bitten by these problems, like, less-than-once-per-career levels of rare (if you honor the advice above)... but certainly not unheard of, definitely not a zero base rate.