Improvements to CLI usability

[brainless]

Since are building our crawling to detect text templates or HTML node duplicates across a domain, let us separate the preparation and the crawler stages in the CLI.

Tasks

• Change CLI argument --link to --domain where a user can pass a URL or domain without protocol or path
• Extract domain name from --domain argument, trim
• Create a CLI argument --prep, remove --verbose and --template arguments
• Add tests for these changes, update tests for features being removed

CLI argument --prep

• In this mode, we will crawl up to 10 pages of this domain (instead of current 3)
• Create store, if it does exist, for detected templates and element's path they were found at
• Element's path should be from HTML root of the page and should contain tag name and class list for each element in the path (no id)
• Run the template detection logic across all crawled pages
• Print all detected paths, de-duplicated from all pages at the end
• The print out should be a serialized format that can be used by Rust