[VectorDistribute] Rework LDS operand promotion (#24408)

Rework how promotion of operands to LDS works in the VectorDistribute
pipeline.

So far, `linalg.copy` operations were inserted early in the pipeline.
Now, we skip the insertion of `linalg.copy` for operands (we keep the
behavior for results). Instead, the analysis from
https://github.com/iree-org/iree/pull/24227 propagates the promotion
types upwards from the compute operations that actually configure the
promotion of operands up to the operations accessing the data in memory
(`transfer_read`, `gather`) when we reach `GPUVectorAlloc`. Based on the
propagated information, the necessary promotion to LDS can be inserted.

Layout conflicts are still resolved through an LDS roundtrip at the
conflict point.

Assisted-by: Codex

---------

Signed-off-by: Lukas Sommer <lukas.sommer@amd.com>