lima/ppir: enable vectorize optimization
pp has vector units and some operations can be optimized when bundled together. Benchmarking this with piglit shaders shows that the instruction count can be greatly reduced on many examples with vectorize.
This also exposed an issue with lima receiving vector select from nir. nir vec4 fcsel seems to assume that each component of the condition will be used to select the same component from the options, but pp can't implement that since it only has 1 component for the condition.