[Bug tree-optimization/126706] New: `(int)(float)(int)(unsigned short)a` should be optimized to `(int)(unsigned short)a`

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126706

            Bug ID: 126706
           Summary: `(int)(float)(int)(unsigned short)a` should be
                    optimized to `(int)(unsigned short)a`
           Product: gcc
           Version: 17.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---
            Target: aarch64 x86_64

Take:
```
_BitInt(32) f(unsigned _BitInt(16) t1)
{
  _BitInt(32) t = t1;
  _Float32 tt = t;
  return tt;
}
```

This should be optimized to just return t1; as the range of t fits into float.
Note I used C's _BitInt here so I can be explict about the precision.

That is this:
```
    /* If we are converting an integer to a floating-point that can
       represent it exactly and back to an integer, we can skip the
       floating-point conversion.  */
    (if (GIMPLE /* PR66211 */
         && inside_int && inter_float && final_int &&
         (unsigned) significand_size (TYPE_MODE (inter_type))
         >= inside_prec - !inside_unsignedp)
     (convert @0)))))))

```

Could use some range info