[mlir] Fixed typo in type (128x64 -> 64x128) in TMA load test (#70022)

The test was meant to check `64x128xf16` as the contiguous dimension
exceeds the cache line (128b). TMA requires cache line-aligned loads, so
loading 64x128 can be done with two 64x64 loads, as documented in the
test.

However, there was a typo in the type, which was `memref<128x64xf16>`
instead of the correct `memref<64x128xf16>`. This PR corrects the issue
and updates the verification.

Author: grypp