Bug 1839487 Add new suggested_index columns to quicksuggest

[0c0w3]

https://bugzilla.mozilla.org/show_bug.cgi?id=1855338

Checklist for reviewer:

• Commits should reference a bug or github issue, if relevant (if a bug is referenced, the pull request should include the bug number in the title)
• If adding a new field, the field should have a description (see Bug 1652842 - Add a blocklist field so the environment schema works with pre- and post-bug-1647225 data #576 for an example)
• If coming from a fork, run integration tests: ./.github/push-to-trigger-integration <username>:<branchname>

For glean changes:

• Update templates/include/glean/CHANGELOG.md

For modifications to schemas in restricted namespaces (see CODEOWNERS):

• Follow the change control procedure

[0c0w3]

cc @ncloudioj @quiiver, thanks

[ncloudioj]

LGTM!

[quiiver]

What is the reason for having this be non-relative when we already have the global position field?

[0c0w3]

It's helpful to consider a couple different uses of "relative":

1. suggested_index can be relative to the Firefox Suggest group (suggested_index_relative_to_group = true) or to the whole results list (suggested_index_relative_to_group = false), although the latter might also be referred to as "global" or "non-relative"
2. suggested_index can be relative the beginning of the group/list or the end. Positive values are relative to the beginning, negative to the end

In practice suggested_index is helpful when:

• It's relative to the Firefox Suggest group, or
• It's relative to the end of the results list

In contrast, if it's relative to the start of the results list, then it's practically the same as position, modulo the different bases (1-based vs. 0-based). The only time it would be really different is when there are fewer results in the list than the suggested index, e.g. there are only 3 results but the suggested index is 9. But in practice we never use large suggested indexes relative to the whole list. We only ever use a value of 1, for top picks.

All that said, the reason we added these was to support the proposed second treatment branch of higher-placement sponsored suggestions where they're shown at the top of the Suggest section. We would be able to distinguish that case by checking suggested_index = 0 and suggested_index_relative_to_group = true. In general we'd be able to distinguish any other placements we'd want to experiment with. Now that we aren't doing that second branch, at least not right now, we don't actually need these values and so we could just remove them.