refactor: Split monolithic sqlite.go into focused files (bd-0a43)

Split internal/storage/sqlite/sqlite.go (1050 lines) into focused files
for better maintainability and code discovery:

- store.go (306 lines): SQLiteStorage struct, New() constructor,
  initialization logic, and database utilities (Close, Path, IsClosed,
  UnderlyingDB, UnderlyingConn, CheckpointWAL)

- queries.go (1173 lines): Issue CRUD operations including CreateIssue,
  GetIssue, GetIssueByExternalRef, UpdateIssue, UpdateIssueID,
  CloseIssue, DeleteIssue, DeleteIssues, SearchIssues with all helpers

- config.go (95 lines): Configuration and metadata management (SetConfig,
  GetConfig, GetAllConfig, DeleteConfig, SetMetadata, GetMetadata) plus
  OrphanHandling type definitions

- comments.go (83 lines): Comment operations (AddIssueComment,
  GetIssueComments)

- sqlite.go (31 lines): Package documentation explaining file organization

Additional changes:
- Removed duplicate OrphanHandling definition from ids.go (was causing
  build error with new config.go)

Impact:
- Zero functional changes, all tests pass (2.6s runtime)
- Improved code discovery: easy to locate specific functionality
- Better maintainability: related code grouped logically
- Reduced cognitive load: smaller, focused files

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

Co-Authored-By: Claude <noreply@anthropic.com>

internal/storage/sqlite/ids.go

@@ -183,16 +183,6 @@ func tryResurrectParent(parentID string, issues []*types.Issue) bool {
 	return false // Parent not in this batch
 }
 
-// OrphanHandling defines how to handle missing parent issues during import
-type OrphanHandling string
-
-const (
-	OrphanStrict     OrphanHandling = "strict"     // Fail import on missing parent
-	OrphanResurrect  OrphanHandling = "resurrect"  // Auto-resurrect from batch
-	OrphanSkip       OrphanHandling = "skip"       // Skip orphaned issues
-	OrphanAllow      OrphanHandling = "allow"      // Allow orphans (default)
-)
-
 // EnsureIDs generates or validates IDs for issues
 // For issues with empty IDs, generates unique hash-based IDs
 // For issues with existing IDs, validates they match the prefix and parent exists (if hierarchical)