We do use the host FPU for a subset of floating point operations now. However it only really works for clean 32/64 IEEE FP so anything that goes through the x87 still needs software emulation.
(I might also look at the other issues that I have with TCG code quality sometime)
Might be worth it to update the wiki too.
I made a Qemu backend targeting LLVM before, but that turned out to be way too heavy to be usable, I wonder if it's worth revisiting that idea nowadays...