Remove ExpInt from Float, always use i32 instead

`ExpInt` is likely to only have performance benefits on 16-bit
platforms, but makes working with the exponent more difficult. It seems
like a worthwhile tradeoff to instead just use `i32`, so do that here.

library/compiler-builtins/libm/crates/libm-test/src/f8_impl.rs

@@ -20,7 +20,6 @@ pub struct f8(u8);
 impl Float for f8 {
     type Int = u8;
     type SignedInt = i8;
-    type ExpInt = i8;
 
     const ZERO: Self = Self(0b0_0000_000);
     const NEG_ZERO: Self = Self(0b1_0000_000);
@@ -62,10 +61,6 @@ impl Float for f8 {
         self.0 & Self::SIGN_MASK != 0
     }
 
-    fn exp(self) -> Self::ExpInt {
-        unimplemented!()
-    }
-
     fn from_bits(a: Self::Int) -> Self {
         Self(a)
     }