There was this weird bug happening only when `SQLLAB_BACKEND_PERSISTENCE
= False` where if no database had been selected, there would be some
issue.
This makes sure that the store is populated consistently with the list
of database connections regardless of `SQLLAB_BACKEND_PERSISTENCE`. It
also uses the DAO to fetch database and will apply the RBAC-related
filters if any, the same way that the API does.
* feat(reports): scheduler and delivery system
* working version
* improvements and fix grace_period
* add tests and fix bugs
* fix report API test
* test MySQL test fail
* delete-orphans
* fix MySQL tests
* address comments
* lint
* feat: new report schedule models
* lint and unique constraint
* support sqlite
* fix sqlite
* add audit mixin and minor fixes
* feat(api): alerts and reports REST API
* feat: new report schedule models
* lint and unique constraint
* support sqlite
* fix sqlite
* add audit mixin and minor fixes
* feat(api): alerts and reports REST API
* draft working version
* add tests
* test
* black
* remove copy pasta
* solve dashboard object representation being used on cache
* tests and custom filter
* fix PUT has PATCH on active field
* create feature flag
* fix lint
* address comments
* Bump pylint version to 2.5.3
* Add a global disable for the most common new pylint error
* Fix a bunch of files containing very few errors
* More pylint tweakage, low-hanging fruit
* More easy stuff...
* Fix more erroring files
* Fix the last couple of errors, clean pylint!
* Black
* Fix mypy issue in connectors/druid/models.py
* [dashboard] Refactor API using SIP-35
* [dashboard] Fix, import
* [dashboard] more tests
* [dashboards] a misc of improvements
* [charts] Fix, DAO and tests
* [dashboards] small exceptions refactor
* [dashboards] lint
* [dashboards] Improves comments on base classes
* [dashboards] lint
