RichEditor migration from v3 to v4: Image captions lost due to TipTap editor change

Hi, after upgrading from Filament v3 to v4, I'm experiencing critical data loss with existing rich text content that contains images with captions. The problem stems from the RichEditor component switching to TipTap as the underlying editor. The Problem: Existing Content Structure (Trix/v3): Images with captions were stored using semantic HTML with <figure> and <figcaption> tags (see the image attached that shows before and after saving) Data Corruption After Migration (TipTap/v4): When content is saved in the new TipTap editor, the structure is automatically corrupted The <figcaption> tags are stripped and converted to regular <p> tags Caption text is completely lost or converted to separate paragraph text (see the image that shows before and after saving) Missing Caption Functionality: The new TipTap-based RichEditor image insert only provides an "alt text" field for images There's no UI to add or edit captions (figcaptions) for images How can i solve these issues? Thank you