Comment out top logo div and enable Flash-DMA banner in README and README_zh

Author: LoserCheems

README.md

@@ -1,6 +1,6 @@
-<div align="center">
+<!-- <div align="center">
   <img src="./assets/logo.png" alt="SmallDoges" width="100%">
-</div>
+</div> -->
 
 <div align="center">
 
@@ -10,7 +10,7 @@
 </div>
 
 
-<!-- ![Flash-DMA Banner](assets/flash_dmattn_banner.png) -->
+![Flash-DMA Banner](assets/flash_dmattn_banner.png)
 
 Flash-DMA is a high-performance attention implementation that integrates Flash Attention's memory efficiency with Dynamic Mask Attention's sparse computation capabilities for processing extremely long sequences in transformer models.
 

README_zh.md

@@ -1,6 +1,6 @@
-<div align="center">
+<!-- <div align="center">
   <img src="./assets/logo.png" alt="SmallDoges" width="100%">
-</div>
+</div> -->
 
 <div align="center">
 
@@ -10,7 +10,7 @@
 </div>
 
 
-<!-- ![Flash-DMA Banner](assets/flash_dmattn_banner.png) -->
+![Flash-DMA Banner](assets/flash_dmattn_banner.png)
 
 Flash-DMA 是一个高性能的注意力实现，将 Flash Attention 的内存效率与动态掩码注意力的稀疏计算能力相结合，用于在 Transformer 模型中处理超长序列。