Update on "[ET-VK] Change weight packing in embedding"

The existing weight tensor for aten.embedding is created using a `tensor_like` from the output tensor, which defaults to channel packed.

However, the weight tensor is actually a 2D-tensor of `(num_embedding, dim_of_embedding)`. It is better in space to use either width or height packing.

This diff changes the implementation to use height-packing.

Differential Revision: [D66421366](https://our.internmc.facebook.com/intern/diff/D66421366/)

[ghstack-poisoned]

Author: yipjustin

backends/vulkan/runtime/graph/ops/impl/Embedding.cpp

@@ -19,8 +19,8 @@
 
 namespace vkcompute {
 
-using utils::StorageType;
 using utils::GPUMemoryLayout;
+using utils::StorageType;
 
 void check_embedding_args(
     const api::vTensor& weight,