Harden TechJournalist media bootstrap per PR #537 review feedback

[Copilot]

Bug Fix

What was the bug?

This PR addresses all actionable comments in the linked PR #537 review thread for the Phi-4 TechJournalist notebook. The notebook’s media bootstrap path lacked bounded download/error handling, and image loading could retain an open file handle.

How did you fix it?

• Media download reliability

  • Added Path("media") bootstrap and standardized image_path / audio_path.
  • Replaced unbounded retrieval with a timeout-bound download path.
  • Wrapped download failures with a clear, actionable RuntimeError instructing manual placement under media/.

• Resource handling for image input

  • Updated image loading to use a context manager and copy the image into memory to avoid lingering file handles.

• Notebook wiring updates

  • Added urllib.request and Path imports required by the new bootstrap logic.
  • Switched audio/image reads from hardcoded local files to the new media paths.

for path, url in sample_media.items():
    if not path.exists():
        try:
            with urllib.request.urlopen(url, timeout=30) as response, open(path, "wb") as out_file:
                out_file.write(response.read())
        except Exception as exc:
            raise RuntimeError(
                f"Failed to download {path.name} from {url}. Download it manually and place it under {media_dir}/"
            ) from exc

with Image.open(image_path) as img:
    image = img.copy()

Testing

Scope is limited to the requested review-thread fixes in the notebook.