[llvm] [LangRef] Clarify specification for float min/max operations (PR #172012)

[Ralf Jung via llvm-commits]

RalfJung wrote:

This sounds good to me! I don't have a strong opinion on whether `minnum` should default to `nsz` behavior (to match IEEE-754) or not.

https://github.com/llvm/llvm-project/pull/172012