[SPO] Fix bugs when handling drive_items in incremental sync

[jedrazb]

Closes https://github.com/elastic/enterprise-search-team/issues/6051

This fixed the issue with changed content of a file (driveItem) not being picked up by an incremental sync.

The simple solution is to disable timestamp checks for site_drives and then rely on efficient delta api to get all changes for drive items.

I verified that the drive_item <-> drive_site is the only scenario when change to an object doesn't update its "parent" lastModifiedDateTime field. Interestingly, drive item updates would result in timestamp updates for its parent "list" and "site", so we are safe at keeping timestamp checks for sites.

For other object types in SPO there were no "unexpected" behaviour when it comes to updating timestamps after edits.

Bonus bug fix

I realized that with fetch_drive_item_permissions enabled, the incremental sync will fail at handling DELETE operations since we are trying to fetch ACLs for a deleted document - it will result in 404 error. I added a logic to handle that scenario and manually tested that it works.

Checklists

Pre-Review Checklist

• this PR has a meaningful title
• this PR links to all relevant github issues that it fixes or partially addresses
• if there is no GH issue, please create it. Each PR should have a link to an issue
• this PR has a thorough description
• Covered the changes with automated tests
• Tested the changes locally
• Added a label for each target release version (example: v7.13.2, v7.14.0, v8.0.0)
• Considered corresponding documentation changes
• Contributed any configuration settings changes to the configuration reference

[github-actions]

💚 Backport PR(s) successfully created

Status	Branch	Result
✅	pre-8.10-stable	#1828
✅	8.11	#1829

The backport PRs will be merged automatically after passing CI.

[seanstory]

💚 All backports created successfully

Status	Branch	Result
✅	8.10

Questions ?

Please refer to the Backport tool documentation