PQ preprocessor does not natively support padded datasets

[aamijar]

Currently, the make_device_pq_dataset can accept a padded_dataset type but internally copies it to a contiguous dense dataset since its helpers like train_from_rows do not support encoding padded datasets.

cuvs/cpp/src/preprocessing/quantize/pq.cu

Lines 99 to 111 in 51dc5d4

	if (stride != dim) {
	auto dense = raft::make_device_matrix<T, int64_t>(res, n_rows, dim);
	raft::copy_matrix(dense.data_handle(),
	dim,
	device_ptr,
	stride,
	dim,
	n_rows,
	raft::resource::get_cuda_stream(res));
	auto dense_view =
	raft::make_device_matrix_view<const T, int64_t>(dense.data_handle(), n_rows, dim);
	return detail::vpq_build_half(res, params, dense_view);
	}