[llvm] [LoongArch] Use LSX for scalar FP rounding with explicit rounding mode (PR #114766)

[via llvm-commits]

zhaoqi5 wrote:

> LGTM. But it would be better precommit the new test before this PR.

@SixWeining . A test was added in https://github.com/llvm/llvm-project/pull/114968. Thanks.

https://github.com/llvm/llvm-project/pull/114766