Define UB in float-to-int casts to saturate

- Round to zero, and representable values cast directly.
- `NaN` goes to 0
- Values beyond the limits of the type are saturated to the "nearest value"
  (essentially rounding to zero, in some sense) in the integral type, so e.g.
  `f32::INFINITY` would go to `{u,i}N::MAX.`

src/test/ui/numbers-arithmetic/saturating-float-casts.rs

@@ -1,6 +1,5 @@
 // run-pass
 // Tests saturating float->int casts. See u128-as-f32.rs for the opposite direction.
-// compile-flags: -Z saturating-float-casts
 
 #![feature(test, stmt_expr_attributes)]
 #![deny(overflowing_literals)]