It a little tricker than that. Both Rust and C++ can auto-execute code at the end of the scope of a block of code (stack level destructors). So you would still be imposing extra work on the programmer to use a Rust back-end exposed through a TypeScript front-end, and you'd end up with an exposed API exactly the same as the current TypeScript version.
However if the full application model is all done in Rust you avoid that extra burden of the missing stack level destructors. Yet you'd still be back-paddling a bit with interop between Rust and JavaScript. It is not a very productive work-flow, you'd only push in that direction if your desperate for high performance in the browser.
I'll still pursue the Rust implementation, mainly because I've already invested the time/effort in learning it real well. And I find it an interesting language. As for work projects where productivity is important, Kotlin looks more attractive to me if your existing code-base exists on the jvm.