Use floats for wasmtime::component::Val::Float* (#5510)

The definitions of `wasmtime::component::Val::Float{32,64}` mirrored
`wasmtime::Val::F{32,64}` by using integers as their wrapped types,
storing the bit representation of their floating point values.
This was necessary for the core Wasm `f32`/`f64` types because Rust
floats don't have guaranteed NaN bit representations.

The component model `float32`/`float64` types require NaN
canonicalization, so we can use normal Rust `f{32,64}` instead.

Closes #5480

crates/fuzzing/src/generators/component_types.rs

@@ -34,8 +34,8 @@ pub fn arbitrary_val(ty: &component::Type, input: &mut Unstructured) -> arbitrar
         Type::U32 => Val::U32(input.arbitrary()?),
         Type::S64 => Val::S64(input.arbitrary()?),
         Type::U64 => Val::U64(input.arbitrary()?),
-        Type::Float32 => Val::Float32(input.arbitrary::<f32>()?.to_bits()),
-        Type::Float64 => Val::Float64(input.arbitrary::<f64>()?.to_bits()),
+        Type::Float32 => Val::Float32(input.arbitrary()?),
+        Type::Float64 => Val::Float64(input.arbitrary()?),
         Type::Char => Val::Char(input.arbitrary()?),
         Type::String => Val::String(input.arbitrary()?),
         Type::List(list) => {