Update on "[ET-VK][ez] Allow logit linear layer to be lowered to Vulkan"

## Context

Due to poor performance of Vulkan's int4 linear operator, the final logit layer of the transformer model was not being delegated to vulkan, and was instead quantized and executed with the XNNPACK delegate.

However, with D72412950 / #9883 decent performance can now be achieved with Vulkan/s int4 linear op. Therefore, the final logit layer can be lowered to Vulkan.

## Changes

* Remove limit from `VkInt4WeightOnlyQuantizer` that was causing it to ignore the logit layer of the transformer
* Do not apply XNNPACK partitioner and quantizer when lowering with Vulkan

Differential Revision: [D72480177](https://our.internmc.facebook.com/intern/diff/D72480177/)

cc manuelcandales cbilgin

[ghstack-poisoned]

Author: SS-JIA

backends/vulkan/_passes/squeeze_unsqueeze_inputs.py

@@ -30,6 +30,12 @@ class SqueezeUnsqueezeInputs(ExportPass):
     def should_squeeze(self, op, shape: List[int]) -> bool:
         if len(shape) == 3:
             return shape[1] == 1 and shape[0] > 1
+        if len(shape) == 4:
+            # No need to squeeze if all dims are 1 except the width dim
+            if all(dim == 1 for dim in shape[:-1]):
+                return False
+            # Otherwise, check for squeezable dim
+            return 1 in shape[:-1]
 
         # Prefer not to introduce additional orchestration ops by default
         return False