Add weight layout transformation cache for Conv operator

[jchen10]

Implement lazy weight layout transformation for WebGPU Conv kernel to avoid redundant GPU transposes on every inference.

Key changes:

• Add WeightLayoutTransformCache to cache transformed weights by name and format
• Implement TransformWeightLayout() helper using existing TransposeKernel for OIHW->HWIO transformation
• Cache stored in WebGpuExecutionProvider, shared across all kernels

[jchen10]

Follow-up for #26554

[jchen10]

@fs-eire PTAL

[jchen10]

I am still looking into the PrePack approach, which seems more appealing as it does release the original tensors.

[fs-eire]

I am still looking into the PrePack approach, which seems more appealing as it does release the original tensors.

Please take a look at #26602. However I didn't finish all validation yet.

[jchen10]

I am still looking into the PrePack approach, which seems more appealing as it does release the original tensors.

Please take a look at #26602. However I didn't finish all validation yet.

Great. That's exactly what I wanted. One pity of PrePack is that we couldn't know the runtime input/output shapes which may impact how we choose the optimal blocked format for weight. Let's see if this issue will come up in the future. So far so good.