feat(editor): add Bear backup import and markdown zip folder hierarchy (#14599)

## Summary

- Add Bear `.bear2bk` backup importer (TextBundle-based zip format)
- Enhance markdown zip import to preserve folder structure from zip
paths
- Add colored highlight (`<mark data-color="...">`) support to HTML
adapter

### Bear Import Details

Bear backups are zip archives of TextBundle directories. The importer:
- Parses Bear-specific markdown (highlights `==text==`, callouts `>
[!NOTE]`, inline tags `#tag`)
- Extracts creation/modification dates from `info.json` metadata
- Filters out trashed notes
- Converts Bear tags to AFFiNE tags (consolidated by root segment)
- Builds folder hierarchy from nested tag paths (e.g.,
`#work/projects/alpha`)
- Uses JSZip for lazy decompression to handle large backups without OOM

### Markdown Zip Folder Hierarchy

`importMarkdownZip` now returns `{ docIds, folderHierarchy }` instead of
just `docIds[]`, enabling the UI to recreate the zip's directory
structure as AFFiNE folders.

## Related Issues

- Implements the TextBundle-based import approach suggested in #14115 /
Discussion #14142
- Addresses folder structure preservation requested in #10003
- Partially addresses frontmatter metadata import from #11286

## Test Plan

- [ ] Import a Bear `.bear2bk` backup file via the import dialog
- [ ] Verify tags are created and assigned to documents
- [ ] Verify folder hierarchy matches Bear's nested tag structure
- [ ] Verify creation/modification dates are preserved
- [ ] Verify highlighted text and callouts render correctly
- [ ] Verify images and attachments are imported
- [ ] Import a markdown zip with nested folders, verify folder structure
is recreated
- [ ] Verify trashed Bear notes are excluded

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **New Features**
* Bear (.bear2bk) backup import: bulk import notes, convert/dedupe tags,
create nested folders, and return imported doc IDs plus folder
hierarchy; UI import option and progress integrated.
* Markdown ZIP import now returns an optional folder hierarchy alongside
created doc IDs.

* **Bug Fixes / Improvements**
* Highlighting: mark elements validate color names, default safely, and
apply consistent background styling.

* **Chores**
  * Added runtime dependency for ZIP handling.

* **Documentation**
  * Added localization strings and i18n accessors for Bear import UI.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: DarkSky <25152247+darkskygit@users.noreply.github.com>

blocksuite/affine/inlines/preset/src/adapters/html/html-inline.ts

@@ -320,9 +320,21 @@ export const htmlMarkElementToDeltaMatcher = HtmlASTToDeltaExtension({
     if (!isElement(ast)) {
       return [];
     }
+    const dataColor =
+      typeof ast.properties?.dataColor === 'string'
+        ? ast.properties.dataColor
+        : '';
+    const colorName =
+      dataColor &&
+      /^(red|orange|yellow|green|teal|blue|purple|grey)$/.test(dataColor)
+        ? dataColor
+        : 'yellow';
     return ast.children.flatMap(child =>
       context.toDelta(child, { trim: false }).map(delta => {
-        delta.attributes = { ...delta.attributes };
+        delta.attributes = {
+          ...delta.attributes,
+          background: `var(--affine-text-highlight-${colorName})`,
+        };
         return delta;
       })
     );