Backend.ai-manager

Fixes
* Add missing login session token (AIOHTTP_SESSION) in initiating app proxying. ([594](https://github.com/lablup/backend.ai-manager/issues/594))
* Bump `backend.ai-common` to 22.03.1.

Features
* Wrap `UnknownImageReference` as `ObjectNotFound` error instead of `InternalServerError` ([591](https://github.com/lablup/backend.ai-manager/issues/591))

Fixes
* Fix to mount dotfolder when creating a session, which was omitted when there was no vfolder to explicitly mount. ([588](https://github.com/lablup/backend.ai-manager/issues/588))
* Change the default values of `dependencies`, `callback_url` and `startup_command` to `None` instead of `undefined` (which is for template arguments). ([589](https://github.com/lablup/backend.ai-manager/issues/589))
* Temporarily rollback the default distributed lock implementation to `pg_advisory` from `etcd` until we identify and fix spurious lock timeout issues ([592](https://github.com/lablup/backend.ai-manager/issues/592))
* Update `AbstractDistributedLock` import path to keep compatibility with `backend.ai-common`. ([593](https://github.com/lablup/backend.ai-manager/issues/593))

Features
* Give `group_id` argument when it is needed to query vfolders that allowed by resource policy. ([585](https://github.com/lablup/backend.ai-manager/issues/585))

Fixes
* Do not create a compute session when the requested memory is not larger than the shared memory. ([584](https://github.com/lablup/backend.ai-manager/issues/584))
* Update mypy version to 0.950 and fix newly found type check problems. ([586](https://github.com/lablup/backend.ai-manager/issues/586))

Features
* Add `KeyPairResourcePolicy.max_session_lifetime` to force-terminate sessions after a specified timeout ([583](https://github.com/lablup/backend.ai-manager/issues/583))

Fixes
* Fix a bogus error by skipping database queries with empty data when syncing kernel statistics ([580](https://github.com/lablup/backend.ai-manager/issues/580))
* Remove `GenericNotFound` and split use cases for `URLNotFound` and `ObjectNotFound` for easier debugging and clear error messages ([581](https://github.com/lablup/backend.ai-manager/issues/581))
* All concrete not-found errors (such as `SessionNotFound`, `VFolderNotFound`, etc.) to inherit the `ObjectNotFound` exception ([582](https://github.com/lablup/backend.ai-manager/issues/582))

Fixes
* Remove duplicated logic to obtain redis local time in streaming proxy requests, which caused the depletion of the Redis connection pool. ([576](https://github.com/lablup/backend.ai-manager/issues/576))
* Fix regression of the logic in getting the container stats per period. ([577](https://github.com/lablup/backend.ai-manager/issues/577))
* Make updating scaling group works again by converting the primitive values of the `scheduler_opts` to a `ScalingGroupsOpts` type to pass the trafaret validation step. ([578](https://github.com/lablup/backend.ai-manager/issues/578))
* Remove explicit usage of `pg_advisory_lock` on scheduler. ([579](https://github.com/lablup/backend.ai-manager/issues/579))

Features
* Add an scheduler_option 'pending_timeout' to auto-cancel a pending session after specified timeout ([527](https://github.com/lablup/backend.ai-manager/issues/527))
* Issue BGSAVE or BGREWRITEAOF commands when clearning the session statistics history to ensure compaction and persistence of the Redis database ([565](https://github.com/lablup/backend.ai-manager/issues/565))
* Add the `callback_url` optional parameter to the session creation APIs and call the URL upon session lifecycle events ([567](https://github.com/lablup/backend.ai-manager/issues/567))
* Implement `clean_images` and `modify_image` GraphQL mutation. ([568](https://github.com/lablup/backend.ai-manager/issues/568))
* Make the distributed lock backend replacible, choosing one from "filelock" (for single-node deployments only), "pg_advisory" (current impl.), and "etcd" (future) ([571](https://github.com/lablup/backend.ai-manager/issues/571))

Fixes
* Outer join the user table so that the project vfolder is also counted, and add `user_email` and `group_name` fields to the vfolder's load parameter. ([557](https://github.com/lablup/backend.ai-manager/issues/557))
* Fetch image alias from resolve method, update query statement and raise explicit error message ([560](https://github.com/lablup/backend.ai-manager/issues/560))
* Fix a regression in the session usage statistics API due to recently changed structure of kernel's mount information in DB ([561](https://github.com/lablup/backend.ai-manager/issues/561))
* Do not call `MSET` to Redis with empty key-value lists when recalculating resource usage ([563](https://github.com/lablup/backend.ai-manager/issues/563))
* Make admins query other users' container logs or destroying them again by explicitly passes the `owner_access_key` parameter to the `get_access_key_scope` function. ([564](https://github.com/lablup/backend.ai-manager/issues/564))
* Fix `inspect_image`, `forget_image` and `alias` commands under `backend.ai mgr etcd` command group failing to run. ([569](https://github.com/lablup/backend.ai-manager/issues/569))
* Migrate `manager.distributed` module to the common pkg and make `PgAdvisoryLokc` a local implementation ([570](https://github.com/lablup/backend.ai-manager/issues/570))
* Migrate the bgtask framework to the common pkg ([572](https://github.com/lablup/backend.ai-manager/issues/572))
* Fix `occupied_slots` column on `agents` kernel does not match to actual resource occupied on corresponding agent. ([574](https://github.com/lablup/backend.ai-manager/issues/574))
* Add explicit lifetime hints for distributed locks used in global timers to guarantee liveness ([575](https://github.com/lablup/backend.ai-manager/issues/575))

Miscellaneous
* Add the migration guide as `MIGRATION.md` ([566](https://github.com/lablup/backend.ai-manager/issues/566))