Remove note about GPU for phi3v (openvinotoolkit#2053)

according to validation status phi3-vision is working on GPU

Author: eaidova

SUPPORTED_MODELS.md

@@ -370,7 +370,6 @@ The pipeline can work with other similar topologies produced by `optimum-intel`
         </ul>
       </td>
       <td>
-          <li>GPU isn't supported</li>
           <li>These models' configs aren't consistent. It's required to override the default <code>eos_token_id</code> with the one from a tokenizer: <code>generation_config.set_eos_token_id(pipe.get_tokenizer().get_eos_token_id())</code>.</li>
       </td>
     </tr>

site/docs/supported-models/index.mdx

@@ -58,7 +58,6 @@ To convert InternVL2 models, `timm` and `einops` are required:
 pip install timm einops
 ```
 #### phi3_v {#phi3_v-notes}
-  - GPU isn't supported
   - Example models' configs aren't consistent. It's required to override the default `eos_token_id` with the one from a tokenizer:
     ```python
     generation_config.set_eos_token_id(pipe.get_tokenizer().get_eos_token_id())