Constant folding for `long->float` produces an incorrect result.

[tannergooding]

Consider the following program:

using System;
using System.Runtime.CompilerServices;

long l = 0x4000_0040_0000_0001L;
Console.WriteLine((float)l);
    
Test1(l);
Test2(l);

[MethodImpl(MethodImplOptions.NoInlining)]
static void Test1(long l) => Console.WriteLine((float)l);

[MethodImpl(MethodImplOptions.NoInlining)]
static void Test2(long l) => Console.WriteLine((float)(double)l);

The expected result is:

4.6116866E+18
4.6116866E+18
4.611686E+18

The actual result is:

4.611686E+18
4.6116866E+18
4.611686E+18

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Consider the following program:

using System;
using System.Runtime.CompilerServices;

long l = 0x4000_0040_0000_0001L;
Console.WriteLine((float)l);
    
Test1(l);
Test2(l);

[MethodImpl(MethodImplOptions.NoInlining)]
static void Test1(long l) => Console.WriteLine((float)l);

[MethodImpl(MethodImplOptions.NoInlining)]
static void Test2(long l) => Console.WriteLine((float)(double)l);

The expected result is:

4.6116866E+18
4.6116866E+18
4.611686E+18

The actual result is:

4.611686E+18
4.6116866E+18
4.611686E+18

Author:	tannergooding
Assignees:	-
Labels:	area-CodeGen-coreclr
Milestone:	-