Add functionality to identify and clean up problematic entries in notes

lib/notes.dart

@@ -6,8 +6,16 @@ class Note {
   late final String displayDate;
   String content;
   String? snippet;
+  bool isProblematic;
+  String problemReason;
 
-  Note({required this.date, required this.content, this.snippet}) {
+  Note({
+    required this.date,
+    required this.content,
+    this.snippet,
+    this.isProblematic = false,
+    this.problemReason = '',
+  }) {
     final dtUtc = DateFormat('yyyy-MM-dd HH:mm:ss').parse(date, true);
     final dtLocal = dtUtc.toLocal();
     displayDate = DateFormat('yyyy-MM-dd HH:mm:ss').format(dtLocal);
@@ -152,3 +160,36 @@ Future<List<Note>> searchNotes(String query) async {
       )
       .toList();
 }
+
+// Find potentially problematic entries based on character distribution
+Future<List<Note>> findProblematicEntries() async {
+  const double maxCharPercentage = 0.7; // If a single char makes up more than 70%, it's suspicious
+  const int minLength = 10; // Only check notes longer than this
+
+  // Simple SQLite query that counts character occurrences using replace
+  final List<Map<String, dynamic>> results = await DB.db.rawQuery('''
+    WITH char_counts AS (
+      SELECT 
+        id,
+        date,
+        content,
+        substr(content, 1, 1) as char,
+        (length(content) - length(replace(content, substr(content, 1, 1), ''))) as char_count,
+        length(content) as total_length,
+        cast(length(content) - length(replace(content, substr(content, 1, 1), '')) as float) / length(content) as percentage
+      FROM notes
+      WHERE length(content) >= ?
+    )
+    SELECT *
+    FROM char_counts
+    WHERE percentage > ?
+    ORDER BY date DESC
+  ''', [minLength, maxCharPercentage]);
+
+  return results.map((row) => Note(
+    date: row['date'] as String,
+    content: row['content'] as String,
+    isProblematic: true,
+    problemReason: 'Character "${row['char']}" makes up ${(row['percentage'] * 100).toStringAsFixed(1)}% of the content',
+  )).toList();
+}