Kamu

Changed
- GraphQL: flows are listed ordered by status and last event time
- Merged two methods(`saveEnvVariable` and `modifyEnvVariable`) from `DatasetEnvVarsMut` info one `upsertEnvVariable`
Fixed
- GQL api flows queries now fetch dataset polling source only once per dataset(and only if Ingest flow type is here)
- Flow trigger status now become disable on flow fail

Changed
- Flight SQL protocol now supports anonymous and bearer token authentication
- The `kamu notebook` command now defaults to `DataFusion` engine for speed, but you can switch to Spark with `--engine spark` argument
- The `kamu notebook` command uses new image based on latest Jupyter and new [`kamu-client-python`](https://github.com/kamu-data/kamu-client-python) library
- The `kamu sql server` command now defaults to `DataFusion` engine with interface changed to use `--engine datafusion/spark`, removing the `--flight-sql` flag
- Examples in `examples/flightsql/python` were updated to new auth and showcasing `kamu` Python library
- Most notebooks in `examples/` directory are using `kamu` Python library with `DataFusion` engine, with Spark still in use for GIS extensions

Fixed
- GraphQL: in a multi-tenant workspace, `datasets.createEmpty` and `datasets.createFromSnapshot` mutations now return dataset aliases prefixed with account name.
- Fix DB transaction error in `/verify` REST endpoint (cherry-picked from `0.214.1`)

Added
- New entity `FlowTrigger` which is now responsible for flow activation and schedules
Changed
- `DatasetFlowConfigsMut` now has only one method `setConfig` for all types of configurations

Added
- New `kamu system decode` command that can decode an arbitrary block file for debugging
- `export` command for bulk data exporting
Changed
- `sql` command now allows to export query command results to file(s)
- FlightSQL session state management improvements

Fixed
- Removed all occurrences of `DataWriterMetadataState` from telemetry spans (too much pressure)