Use stored scraped content for top pages instead of crawling inside canonical audit

**Is your feature request related to a problem? Please describe.**
https://github.com/adobe/spacecat-audit-worker/pull/326/files#diff-88d6eda336d671388349471e2f7210bbb88fba4bd0d4fdbbd7b02cc632aaadc7R147
instead of fetching all the URLs here, we could leverage the scraped content stored in S3, so we don't need to make a request for each audit again (canoincal, JSON-LD, hreflang...)
https://wiki.corp.adobe.com/display/AEMSites/SpaceCat+%7C+Core+%7C+Content+Scraper+and+Processing

**Describe the solution you'd like**
Use stored scraped content for top pages instead of crawling inside canonical audit

**Additional context**
[Add any other context or screenshots about the feature request here.
](https://github.com/adobe/spacecat-audit-worker/pull/326)

Start with the canonical audit URL-fetching logic referenced in PR #326 and review the SpaceCat Content Scraper and Processing documentation. Trace how scraped content is stored in S3 and how the canonical audit currently fetches top-page URLs; done means the audit uses the stored content instead of making those repeated requests.

Written by the indexing model from the issue text.