The new option works like the existing LLVM_TABLEGEN, and LLVM_CONFIG_PATH options.
Instead of building llvm-nm, the build uses the executable defined by LLVM_NM.
This is useful for cross-compilation scenarios where the host cannot run the cross-compiled
tool, and recursing into another cmake build is not an option (due to required DEFINE's, for example).
I guess this works and is a no-op, but setting this variable to empty (or not setting it at all) should also work, I believe, and perhaps expresses the intent clearer (that there's no target for llvm-nm and you're just supplying a prebuilt one).