feat: Document-centric scan results + DSI deduplication

DSI Dedup (consent-tester):
- Only H1/H2 headings count as documents (not H3/H4 sub-sections)
- Sub-sections (Cookies, Betroffenenrechte, Social Media) are part of
  parent document's full text, not separate documents
- Reduces IHK result from 30 to ~11 real documents

Backend (agent_scan_routes):
- ScanFinding gets doc_title field linking each finding to its document
- doc_title set when creating DSI findings for document attribution

Frontend (ScanResult.tsx):
- 3 sections: Services table, Document cards, General findings
- Documents: expandable cards with completeness bar (green/yellow/red)
- Findings grouped under their parent document
- Each card shows: title, word count, findings count, % completeness
- Findings without doc_title go to "Allgemeine Findings" section

Email Summary (agent_scan_helpers):
- Findings listed under their parent document
- General findings in separate section
- No more flat mixed list

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

consent-tester/services/dsi_discovery.py

@@ -444,13 +444,18 @@ async def _expand_all_interactive(page: Page) -> None:
 
 
 async def _find_inline_dsi_sections(page: Page) -> list[dict]:
-    """Find DSI content already visible on the page (e.g. expanded accordions)."""
+    """Find DSI content already visible on the page (e.g. expanded accordions).
+
+    Only counts top-level documents (H1/H2 with DSI keywords).
+    Sub-sections (H3/H4 like 'Cookies', 'Betroffenenrechte') are NOT counted
+    as separate documents — their text is part of the parent document.
+    """
     try:
         sections = await page.evaluate("""
             () => {
                 const results = [];
-                // Find headings that match DSI keywords
-                const headings = document.querySelectorAll('h1, h2, h3, h4, h5');
+                // Only H1 and H2 count as document-level headings
+                const headings = document.querySelectorAll('h1, h2');
                 const dsiKeywords = [
                     'datenschutz', 'privacy', 'données', 'privacidad', 'protezione',
                     'gegevensbescherming', 'ochrona danych', 'tietosuoja', 'integritet',
@@ -461,12 +466,13 @@ async def _find_inline_dsi_sections(page: Page) -> list[dict]:
                     const textLower = text.toLowerCase();
                     if (!dsiKeywords.some(kw => textLower.includes(kw))) continue;
 
-                    // Get the section content following this heading
+                    // Get ALL content until the next H1/H2 (include sub-sections H3-H5)
                     let content = '';
                     let el = h.nextElementSibling;
                     let count = 0;
-                    while (el && count < 50) {
-                        if (el.tagName.match(/^H[1-5]$/)) break;
+                    while (el && count < 200) {
+                        // Stop at next H1 or H2 (next top-level document)
+                        if (el.tagName === 'H1' || el.tagName === 'H2') break;
                         content += (el.textContent || '').trim() + '\\n';
                         el = el.nextElementSibling;
                         count++;