It seems like the first step should be packaging this up as a metapackage or something for interested parties and having at least a full unstable release cycle where that's optional and tested before replacing such utilities for what is oestensibly an end-user distribution for less-technical people who will not be able to fix things going wrong deep in the heart of shell scripts.

I'm all for Rusting more-or-less all the things, but Rust isn't actually magic or anything. Things aren't automatically better just because Rust. And as someone who keeps at least modest track of CVEs and such, these utilities aren't exactly throwing CVEs out left right front and center. I don't like the amount of C in the world but being run billions of times a day in every conceivable environment is itself a really, really good test. We well know from history, still not always perfect, but honestly these aren't the things that need a priority Rusting; it's the code that doesn't fit that description that really needs it.