Management Command for async script to make all slugs unique

[hannaseithe]

Short description

This PR is based on: #3917

• a (@shared_task) script that makes sure all latest translations of events, pois and pages have unique slugs (EDIT: I have decided to delete the query for only latest translations, since the database constraint will guarantee it for all)
• a management command that queues the script with celery

Proposed changes

• the script is optimized for speed compared to the original migration script
• if this command runs fast enough, I will turn it into a migration and add it to PoC: Add an Exclusion Constraint for pagetranslation's slug uniqueness #3826
• I have added no tests for now, as its unsure we will keep the command anyways and because I see no way to create a test state with duplicate slugs with the application level constraints already in place and the database constraints coming up soon

Side effects

• None

Faithfulness to issue description and design

There are no intended deviations from the issue and design.

How to test

• You will have to manipulate the test-data.json to get duplicate slugs (within one region and language!)
• Or enter the slugs into the database directly (django-admin wont work)
• in the shell you can use the .update method or the bulk_create method as they circumvent the application level constraints that have already been implemented

Then run the command with tools/integreat-cms-cli make_slugs_unique

Resolved issues

Fixes: #
Blocks: #3826
Relates to #3917

---

Pull Request Review Guidelines

[dkehne]

Review Summary

What it does

• Creates a Celery @shared_task called make_all_slugs_unique that iterates through all page/event/POI translations and regenerates slugs to ensure uniqueness
• Adds a management command make_slugs_unique that queues this task
• Supports --dry-run flag and filtering by object type (--objects page event poi)

Key observations

1. Uses existing generate_unique_slug - Reuses the established slug generation logic, which is good for consistency

2. Transaction handling - Uses transaction.atomic() to batch changes, good for data integrity

3. Celery task with .delay() - The command queues the task asynchronously, which is appropriate for potentially long-running operations

4. Missing from __future__ import annotations - The new management command file is missing this required import (per project guidelines)

5. Type import issue - Final is imported inside TYPE_CHECKING block but used at module level for DEFAULT_OBJECTS. This will cause a runtime NameError:

   DEFAULT_OBJECTS: Final[tuple[SlugObject, ...]] = ("event", "poi", "page")

   Final needs to be imported outside the TYPE_CHECKING block.

6. No tests - Author acknowledges this and explains why (can't easily create duplicate slug state with constraints in place)

7. Dry-run passed to Celery - The dry_run parameter is passed to the async task, which works but means you can't see real-time output of what would change

Potential bug: The Final type annotation will cause a NameError at runtime since it's only imported under TYPE_CHECKING.