[webgpu] Optimize DP4AMatMulNBitsSmallMProgram for intel (microsoft#25192)

### Description
This PR optimizes the Intel GPU path for the
`DP4AMatMulNBitsSmallMProgram` by tuning `tile_size` and
`tile_size_k_vec`.



### Motivation and Context
With this change, we achieved >8% performance boost on Intel iGPUs
(Xe-LP and Xe2-LPG) for phi-4-mini-accuracy4 model.

Author: jing-bao

onnxruntime/contrib_ops/webgpu/quantization/dp4a_matmul_nbits.cc

@@ -596,6 +596,11 @@ Status ApplyDP4AMatrixMatMulNBits(const Tensor* a, const Tensor* b, const Tensor
     uint32_t tile_size_k_vec = 16;
     uint32_t tile_size = 32;
 
+    if (context.AdapterInfo().vendor == std::string_view{"intel"}) {
+      tile_size_k_vec = 32;
+      tile_size = 4;
+    }
+
     DP4AMatMulNBitsSmallMProgram mul_program{tile_size_k_vec, tile_size, nbits, has_zero_points};
     uint32_t num_N_tile = (N + tile_size - 1) / tile_size;
     mul_program.SetWorkgroupSize(128);