Fixes #6880

When upserting new points, the is_empty filter did not work properly. It did not always include the newly upserted points during filtering.

More specifically, this problem appears when upserting points in which the payload field (of the index) is absent. The null index relies on knowing the exact (segment) point count to reliably filter points. When upserting a point in which the payload field is absent, we don't add the point to the index and therefore don't bump the point count.

When the payload field is absent, instead of adding (set, overwrite) the point to the index, we actually remove (set, overwrite) it from the index. To fix the filtering problem, I now also bump the point count when removing a point. This will ensure that new upsertions without the field still set the correct point count.

This PR fixes two problems:

1. not setting the total point count (described above)
2. loading the null index from the wrong path (in some scenarios)

I also add a test to assert correct behavior.

Existing deployments do not have permanent damage because of this bug. All (existing) null indices are fixed on restart.

All Submissions:

• Contributions should target the dev branch. Did you create your branch from dev?
• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?

New Feature Submissions:

1. Does your submission pass tests?
2. Have you formatted your code locally using cargo +nightly fmt --all command prior to submission?
3. Have you checked your code using cargo clippy --all --all-features command?