Build a BFS Web Crawler

Implement a web crawler for a single website.

You are given:

• A starting URL startUrl
• A function fetch(url) that returns the HTML content of a page or raises an error
• A function extractLinks(html) that returns all links found on the page

Write a crawler that:

1. Traverses pages in breadth-first order starting from startUrl.
2. Only visits pages in the same domain as startUrl.
3. Normalizes URLs by resolving relative paths, removing fragments, and treating equivalent URLs as the same page.
4. Retries transient fetch failures up to 3 times before giving up.
5. Skips permanently failed pages.
6. Never visits the same normalized URL more than once.

Return the list of visited URLs in BFS order.

In your implementation, be prepared to explain your queue and visited-set design, how you detect duplicate URLs, and how retry logic interacts with the crawl loop.