In siteUtil.updateSiteFavicon use location cache

[ayumi]

Fix #9276

Note: This slightly changes the function behavior because it removes use of normalizeUrl when comparing locations. By default the library removes www from URLs:

'https://www.example.com/about.html#test'
=> 'https://example.com/about.html#test'

We had been testing this case, assuming that we had both https://www.brave.com/ and https://brave.com, calling updateSiteFavicon with https://brave.com and expected both sites to be updated.

I don't think this case actually happens (please correct me if I'm wrong), and the siteKey cache does normalize location when setting and getting so it seems safe.

Auditors: @bsclifton @bbondy

Test Plan:

1. load 10K bookmarks
2. Enable bookmarks bar favicons via Preferences > General > "Bookmarks Bar" => "Text and Favicons".
3. Create a bookmark folder.
4. Navigate to https://archive.org
5. Bookmark page.
6. Navigate to https://wikipedia.org
7. Bookmark page into the folder.
8. Examine bookmarks toolbar; there should be favicons for archive and wikipedia.
9. Open the Bookmarks Manager and confirm there are favicons.

Submitter Checklist:

• Submitted a ticket for my issue if one did not already exist.
• Used Github auto-closing keywords in the commit message.
• Added/updated tests for this change (for new code or code which already has tests).
• Ran git rebase -i to squash commits (if needed).
• Tagged reviewers and labelled the pull request as needed.

Test Plan:

Reviewer Checklist:

Tests

• Adequate test coverage exists to prevent regressions
• Tests should be independent and work correctly when run individually or as a suite ref
• New files have MPL2 license header

Related: #8541

[luixxiul]

@ayumi just a head-up: could I know how this issue should solve #8541? I couldn't tell based on the STR in the 1st post.

[ayumi]

@luixxiul I think STR would be to first load 10K bookmarks, then navigate some websites. The STR for #9276 is related.

[bsclifton]

LGTM 😄 One missing test might be to have a magnet link in the list of sites in the test (in case the code changes, it won't regress if the URL is parsed)