Floating-point to integer conversions should saturate on overflow

[tannergooding]

Floating-point to integer conversions should saturate on overflow

[jaredpar]

This would simply entail that NaN -> 0 and overflow is done via saturation, as described above. If appropriate or helpful, the runtime can define a RuntimeFeature flag to help drive this behavior and indicate that the target is expected to saturate.

The team has been dealing with a number of compatibility breaks recently. Some are just bad luck, changes that had intended consquences, etc ... but many are well intended changes like this. The last 6-8 weeks of the compiler team are disproportionately dealing with the ramifications of these breaking changes. It's really hard to be at the end of that see this and be eager to move forward with it.