fix(13): switch to GPT-4o vision — render PDF pages as images for accurate field placement

- extractPdfText now renders each page to JPEG via @napi-rs/canvas + pdfjs-dist (108dpi)
- field-placement.ts sends rendered page images to GPT-4o with vision (detail: high)
- AI can now visually identify underlines, signature blocks, date fields, initials boxes
- System prompt focuses on visual cues (blank lines, boxes) not text pattern matching
- Handles multi-field lines: separate fields for signature blank and date blank on same line

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

teressa-copeland-homes/package-lock.json

@@ -11,6 +11,7 @@
         "@cantoo/pdf-lib": "^2.6.3",
         "@dnd-kit/core": "^6.3.1",
         "@dnd-kit/utilities": "^3.2.2",
+        "@napi-rs/canvas": "^0.1.97",
         "@react-email/components": "^1.0.10",
         "@react-email/render": "^2.0.4",
         "@vercel/blob": "^2.3.1",
@@ -3307,7 +3308,6 @@
       "resolved": "https://registry.npmjs.org/@napi-rs/canvas/-/canvas-0.1.97.tgz",
       "integrity": "sha512-8cFniXvrIEnVwuNSRCW9wirRZbHvrD3JVujdS2P5n5xiJZNZMOZcfOvJ1pb66c7jXMKHHglJEDVJGbm8XWFcXQ==",
       "license": "MIT",
-      "optional": true,
       "workspaces": [
         "e2e/*"
       ],