Clean up output dir in deploy step, update image gen to gpt-image-1.5.

Remove output/ directory after copying PDFs to category folders in the
datasheets-build branch deploy step. Update generate_images.py to use
gpt-image-1.5 model.

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

Author: kwagyeman

.github/workflows/build.yml

@@ -87,6 +87,7 @@ jobs:
             mkdir -p "$dir"
             cp "$f" "$dir/"
           done
+          rm -rf output
           git add cameras/ shields/ sensors/ lenses/ accessories/ *.pdf 2>/dev/null || true
           git commit -m "Build datasheets from ${GITHUB_SHA::8}"
           git push origin datasheets-build --force