[llvm] [RISCV][VLOPT] Add support for 11.13 integer multiply-add instructions (PR #112216)

[Michael Maitland via llvm-commits]

michaelmaitland wrote:

> > I think the VL on the using add instruction can be optimized further, once we add support for optimizing non-vlmax.
> 
> I'm not sure what this means since there are no add instructions in the tests.

I've removed it from the description. Sorry. 

https://github.com/llvm/llvm-project/pull/112216