Collapse only long code blocks in README (#73)

* Initial plan
* Make code blocks collapsible on homepage
* Refine: Keep only long code blocks collapsible

Co-authored-by: anxiangsir <31175974+anxiangsir@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: anxiangsir <31175974+anxiangsir@users.noreply.github.com>

Author: Copilot

README.md

@@ -132,6 +132,9 @@ We train the model on a mixed dataset comprising 740K samples from LLaVA-OneVisi
 
 ### Using AutoModel (Recommended: transformers==4.57.3)
 
+<details>
+<summary>Click to expand code example</summary>
+
 ```python
 from transformers import AutoModel, AutoImageProcessor
 from PIL import Image
@@ -189,8 +192,13 @@ with torch.no_grad():
     outputs = model(video, patch_positions=patch_positions)
 ```
 
+</details>
+
 ### Loading from Source Code  
 
+<details>
+<summary>Click to expand installation and usage code</summary>
+
 ```bash
 git clone https://github.com/EvolvingLMMs-Lab/OneVision-Encoder.git
 cd OneVision-Encoder
@@ -211,6 +219,8 @@ preprocessor = AutoImageProcessor.from_pretrained(
 )
 ```
 
+</details>
+
 ### Codec Input
 
 Add codec-style input documentation for temporal saliency-based patch selection.