[Fix] Add Convolution Vectorization picture.

docs.js

@@ -123,7 +123,8 @@
       titleEn: "Convolution Vectorization",
       desc: "卷积向量化相关文档",
       descEn: "Convolution vectorization documentation",
-      markdownUrl: "docs/compiler/ConvolutionVectorization.md"
+      markdownUrl: "docs/compiler/ConvolutionVectorization.md",
+      bodyOnly: true
     },
     "triton-on-riscv": {
       title: "Triton 的 RISC-V 适配",

docs/compiler/ConvolutionVectorization.en.md

@@ -19,7 +19,7 @@ Let’s first introduce the MLIR dialects and operations that are needed by the
 - `vector.broadcast`: broadcasts a scalar or vector value to an n-D result vector.
 - `vector.fma`: performs fused multiply-add (FMA) on the vector type.
 
-![Graph of the Coefficients Broadcasting Algorithm](./Images/CoefficientsBroadcasting.png)
+![Graph of the Coefficients Broadcasting Algorithm](/docs/compiler/ConvolutionVectorization-CoefficientsBroadcasting.png)
 
 The figure shows the steps of the algorithm:
 - Iterate over each kernel element and broadcast it to vector1.

docs/compiler/ConvolutionVectorization.md

@@ -18,7 +18,7 @@
 - `vector.broadcast`：将标量或向量广播为 n 维结果向量。
 - `vector.fma`：执行向量融合乘加（FMA）运算。
 
-![Graph of the Coefficients Broadcasting Algorithm](./Images/CoefficientsBroadcasting.png)
+![Graph of the Coefficients Broadcasting Algorithm](/docs/compiler/ConvolutionVectorization-CoefficientsBroadcasting.png)
 
 上图展示了算法流程：