Feature: Group/Bookmark/Cluster Files in Pull Requests

## Feature Request: File Grouping/Bookmarking in Pull Requests

### Problem Statement
When reviewing large pull requests with many changed files, it can be difficult to organize and navigate through related files. Users need a way to group, bookmark, or cluster files together for more efficient review workflows.

### Proposed Solution
Implement a feature that allows users to:
- **Group files** by custom categories (e.g., "Frontend Changes", "API Updates", "Tests")
- **Bookmark specific files** for quick access during review
- **Create file clusters** to view related changes together
- **Save grouping preferences** for recurring review patterns

### Use Cases
1. **Large PR Reviews**: Organize 50+ changed files into logical groups
2. **Multi-reviewer Workflows**: Different reviewers can focus on their domain-specific file groups
3. **Iterative Reviews**: Bookmark files that need follow-up or deeper review
4. **Code Navigation**: Quickly jump between related files (e.g., component + test + styles)

### Suggested Implementation
- Drag-and-drop interface for creating groups
- Collapsible file group sections in PR tree view
- Persistent storage of grouping preferences
- Keyboard shortcuts for quick bookmarking
- Visual indicators (colors/icons) for different groups

### Benefits
- Improved review efficiency for large PRs
- Better mental organization of changes
- Reduced cognitive load when reviewing complex changes
- Enhanced collaboration between multiple reviewers

### Additional Considerations
- Integration with existing PR tree view
- Support for both manual and automatic grouping (by file type, directory, etc.)
- Export/import grouping configurations
- Mobile-friendly interface for grouping on smaller screens

---
🤖 Generated with [Claude Code](https://claude.ai/code)

No files or tests are named; start by locating the existing pull-request tree view and its review-state storage. Clarify the scope and interaction design for grouping, bookmarking, clustering, and persistence, then define completion criteria covering the proposed workflows and any supported shortcuts or responsive behavior.

Written by the indexing model from the issue text.