Fractal-server

* API:
* Add `AccountingRecord` and `AccountingRecordSlurm` tables (\2267).
* Add `/admin/v2/impersonate` endpoint (\2280).
* Replace `_raise_if_naive_datetime` with `AwareDatetime` (\2283).
* Database:
* Add `/admin/v2/accounting/` and `/admin/v2/accounting/slurm/` endpoints (\2267).
* Runner:
* Populate `AccountingRecord` from runner (\2267).
* App:
* Review configuration variables for email-sending (\2269).
* Reduce error-level log records(\2282).
* Testing:
* Drop obsolete files/folders from `tests/data` (\2281).
* Dependencies:
* Bump `httpx` to version `0.28.*` (\2284).

With this release we switch to Pydantic v2.

* Runner:
* Deprecate `FRACTAL_BACKEND_RUNNER="local_experimental"` (\2273).
* Fully replace `clusterfutures` classes with custom ones (\2272).
* Dependencies:
* Bump `pydantic` to v2 (\2270).
* Drop `clusterfutures` dependency (\2272).
* Drop `psutil` dependency (\2273).
* Bump `cryptography` to version `44.0.*` (\2274).
* Bump `sqlmodel` to version `0.0.22` (\2275).
* Bump `packaging` to version `24.*.*` (\2275).
* Bump `cloudpickle` to version `3.1.*` (\2275).
* Bump `uvicorn-workers` to version `0.3.0` (\2275).
* Bump `gunicorn` to version `23.*.*` (\2275).
* Bump `httpx` to version `0.27.*` (\2275).

> Note: this version requires a manual update of email-related configuration variables.

* API:
* Deprecate `use_dataset_filters` query parameter for `/project/{project_id}/dataset/{dataset_id}/images/query/` (\2231).
* App:
* Add fractal-server version to logs (\2228).
* Review configuration variables for email-sending (\2241).
* Database:
* Remove `run_migrations_offline` from `env.py` and make `run_migrations_online` sync (\2239).
* Task lifecycle:
* Reset logger handlers upon success of a background lifecycle operation, to avoid open file descriptors (\2256).
* Runner
* Sudo/SLURM executor checks the fractal-server version using `FRACTAL_SLURM_WORKER_PYTHON` config variable, if set (\2240).
* Add `uname -n` to SLURM submission scripts (\2247).
* Handle `_COMPONENT_KEY_`-related errors in sudo/SLURM executor, to simplify testing (\2245).
* Drop obsolete `SlurmJob.workflow_task_file_prefix` for both SSH/sudo executors (\2245).
* Drop obsolete `keep_pickle_files` attribute from slurm executors (\2246).
* Dependencies:
* Bump `uvicorn` version (\2242).
* Testing:
* Improve testing of sudo-Slurm executor (\2245, \2246).
* Introduce `container` pytest marker (\2249).
* Split CI GitHub Actions in three jobs: API, not API and Containers (\2249).

> WARNING: The database schema update introduced via this version is non-reversible.

* API:
* Drop V1 endpoints (\2230).
* Database:
* Drop V1 tables (\2230).
* Runner:
* Drop V1 runners (\2230).
* Testing:
* Drop V1 tests (\2230).
* Update V2 tests to keep coverage stable (\2230).

* Database
* Drop columns `DatasetV2.filters` and `WorkflowTaskV2.input_filters` (\2232).

This version revamps the filters data structure, and it introduces complex attribute filters.

> Note: This release requires running `fractalctl update-db-data`.
> Some legacy columns will be removed from the database, either as part of
> the `2.11.0` data-migration or as part of the `2.11.1` schema migration.
> Please make sure you have a database dump.

* API:
* Align API with new database schemas for filters-related columns (\2168, \2196, \2202).
* Support importing workflows or datasets with legacy (pre-`2.11.0`) filters-related fields (\2185, \2227).
* Avoid blocking operations from the download-job-logs endpoint, when the zip archive of a running job is requested (\2225).
* Update and simplify `/api/v2/project/{project_id}/status/`, dropping use of temporary job files (\2169).
* Add new (experimental) `/project/{project_id}/workflow/{workflow_id}/type-filters-flow/` endpoint (\2208).
* Database:
* Update table schemas for all filters-related columns:
* Always handle attribute- and type-filters in different columns (\2168).
* Update attribute-filter-values type from scalar to list (\2168, \2196).
* Deprecate attribute filters for `WorkflowTaskV2` (\2168).
* Add attribute filters to `JobV2` (\2168).
* `2.11.0` data-migration script (\2168, \2202, \2208, \2209).
* Runner:
* Introduce database writes in runner component, to replace the use of temporary files (\2169).
* Use `TaskV2.input_types` for filtering, rather than validation (\2191, \2196).
* Make job-execution background-task function sync, to make it transparent that it runs on a thread (\2220).
* Remove all filters from `TaskOutput` (\2190).
* Task Collection:
* Improve logs handling for failed task collections (\2192)
* Testing:
* Speed up CI by splitting it into more jobs (\2210).