Fixup to #13320 .

The requirements for clearing the padding is that it should not be done for views since that would risk clearing valid tensor data and that the tensor memory should be in one contiguous block. The latter is to avoid having to handle the edge case of potentially having to clear memory inside the tensor.

The asserts in mmq.cu and mmvq.cu were too strict because ggml_is_contiguous disallows both views and permutations. I added a new function ggml_is_contiguously_allocated that only disallows views but allows permutations (to my knowledge we do not yet have a utility function for this). I changed the asserts to check that the memory is allocated contiguously and that the tensor in question is not a view of another tensor (to avoid potentially overwriting valid tensor data).

The assert in ggml_mul_mat_id was wrong because I forgot that for MUL_MAT_ID the number of tokens is stored in dimension 2 instead of dimension 1. But since the generic MoE code effectively creates views of src0 I think it's better to mark the slices as such and to remove that assert entirely.

I remembered that in the FlashAttention code the conversion of quantized KV data to FP16 implicitly assumes a contiguous block of memory and added asserts with the new function.