Emphasize typegen as required step after schema changes (databricks#4291

)

## Summary
- Made Step 3 explicit: run typegen after any schema change
- LLMs were adding query schemas but not running `npm run typegen`,
causing TypeScript errors like "query key not assignable to type"

## Test plan
- Documentation change only

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

## Release v0.283.0

### Notable Changes
* Bundle commands now cache the user's account details to improve command latency.
To disable this, set the environment variable DATABRICKS_CACHE_ENABLED to false.

### CLI
* Add commands to pipelines command group ([databricks#4275](databricks#4275))

### Bundles
* Add support for configuring app.yaml options for apps via bundle config ([databricks#4271](databricks#4271))
* Enable caching user identity by default ([databricks#4202](databricks#4202))
* Do not show single node warning when is_single_node option is explicitly set ([databricks#4272](databricks#4272))
* Fix false positive folder permission warnings and make them more actionable ([databricks#4216](databricks#4216))
* Pass additional Azure DevOps system variables ([databricks#4236](databricks#4236))
* Replace Black formatter with Ruff in Python bundle templates for faster, all-in-one linting and formatting ([databricks#4196](databricks#4196))
* engine/direct: support quality monitors ([databricks#4278](databricks#4278))
* Improve folder permissions warning ([databricks#4216](databricks#4216))

### Dependency updates
* Upgrade TF provider to 1.102.0 ([databricks#4235](databricks#4235))
* Upgrade Go SDK to 0.96.0 ([databricks#4206](databricks#4206))

## Release v0.282.0

### Notable Changes
* engine/direct: New plan format (v2) ([databricks#4201](databricks#4201))

### CLI
* Skip non-exportable objects (e.g., `MLFLOW_EXPERIMENT`) during `workspace export-dir` instead of failing ([databricks#4101](databricks#4101))

### Bundles
* Allow `domain_friendly_name` to be used in `name_prefix` in development mode ([databricks#4173](databricks#4173))
* Add missing schema grants privileges ([databricks#4139](databricks#4139))
* Add support for `bundle generate alert` command ([databricks#4108](databricks#4108))
* Add support for `.dbalert.json` files ([databricks#3602](databricks#3602))
* Pass `SYSTEM_TEAMFOUNDATIONCOLLECTIONURI` from env to the Terraform provider for Azure DevOps OIDC auth ([databricks#4169](databricks#4169))
* Add `ipykernel` to the `default` template to enable Databricks Connect notebooks in Cursor/VS Code ([databricks#4164](databricks#4164))
* Add interactive SQL warehouse picker to `default-sql` and `dbt-sql` bundle templates ([databricks#4170](databricks#4170))
* Add `name`, `target` and `mode` fields to the deployment metadata file ([databricks#4180](databricks#4180))
* engine/direct: Fix app deployment failure when app is in `DELETING` state ([databricks#4176](databricks#4176))
* engine/direct: Changes in config that match remote changes no longer trigger an update ([databricks#4201](databricks#4201))

## Release v0.281.0

### CLI
* Fix lakeview publish to default `embed_credentials` to false ([databricks#4066](databricks#4066))

### Bundles
* Add support for configurable catalog/schema for dashboards ([databricks#4130](databricks#4130))
* Pass SYSTEM\_ACCESSTOKEN from env to the Terraform provider ([databricks#4135](databricks#4135))
* `bundle deployment migrate`: when running `bundle plan` propagate `-var` arguments.
* engine/direct: New option --plan to `bundle deploy` to deploy previously saved plan (saved plan with `bundle plan -o json`) ([databricks#4134](databricks#4134))
* engine/direct: Fix dependency-ordered deletion by persisting depends\_on in state ([databricks#4105](databricks#4105))

### Dependency updates
* Upgrade Go SDK to 0.94.0 ([databricks#4148](databricks#4148))
* Upgrade Terraform provider to 1.100.0 ([databricks#4150](databricks#4150))

## Release v0.280.0

### Bundles
* Fixed changing the source for the SQL task from GIT to WORKSPACE ([databricks#4114](databricks#4114))
* `bundle deployment migrate` will now run a plan before migration to check if deployment was done ([databricks#4088](databricks#4088))
* engine/direct: Increase "serial" in state file after every deployment ([databricks#4115](databricks#4115))

### Dependency updates

* Upgrade Go SDK to 0.93.0 ([databricks#4112](databricks#4112))
* Bump Go toolchain to 1.25.5.

## Release v0.279.0

### Notable Changes
* New deployment engine for DABs that does not require Terraform is available to try in experimental mode. Not recommended for production workloads yet. Documentation at [docs/direct.md](docs/direct.md).

### CLI
* Introduce `databricks apps logs` command to tail app logs from the CLI ([databricks#3908](databricks#3908))

### Bundles
* Add support for alerts to DABs ([databricks#4004](databricks#4004))
* Allow `file://` URIs in job libraries to reference runtime filesystem paths (e.g., JARs pre-installed on clusters via init scripts). These paths are no longer treated as local files to upload. ([databricks#3884](databricks#3884))
* Pipeline catalog changes now trigger in-place updates instead of recreation (Terraform provider v1.98.0 behavior change) ([databricks#4082](databricks#4082))

### Dependency updates
* Bump Terraform provider to v1.98.0 ([databricks#4082](databricks#4082))

## Release v0.278.0

### Dependency updates
* Upgrade Go SDK to 0.91.0 ([databricks#3948](databricks#3948))
* Bump Alpine Linux to 3.22 in the Docker image ([databricks#3942](databricks#3942))

### Bundles
* Update templates to use serverless environment version 4 and matching Python version ([databricks#3897](databricks#3897))
* Add a language prompt to the `default-minimal` template ([databricks#3918](databricks#3918))
* Add `default-scala` template for Scala projects with SBT build configuration and example code ([databricks#3906](databricks#3906))

Merge unix and windows SHA files into one (databricks#3915)

## Changes
Merge unix and windows SHA files into one

## Why
The Komac tool (and any other tools expecting a unified checksum file)
will now find the correct checksums in the standard databricks_cli_{{
.Version }}_SHA256SUMS file that contains entries for all platforms.

## Release v0.276.0

### CLI
* Remove previously added flags from the `jobs create` and `pipelines create` commands. ([databricks#3870](databricks#3870))

### Bundles
* Updated the default-python template to follow the Lakeflow conventions: pipelines as source files, pyproject.toml ([databricks#3712](databricks#3712)).
* Fix a permissions bug adding second IS\_OWNER and causing "The job must have exactly one owner." error. Introduced in 0.274.0. ([databricks#3850](databricks#3850))

## Release v0.275.0

### Notable Changes
* Python support for Databricks Asset Bundles is now generally available.

### CLI
* Remove `inplace` mode for the `--progress-format` flag ([databricks#3811](databricks#3811))
* Remove `json` mode for the `--progress-format` flag ([databricks#3812](databricks#3812))
* Deprecate the `--progress-format` flag ([databricks#3819](databricks#3819))

### Bundles
* Add support for `--bind` flag in `bundle generate` ([databricks#3782](databricks#3782))
* Add `pydabs` template replacing `experimental-jobs-as-code` template ([databricks#3806](databricks#3806))
* You can now use the top-level `python` section instead of `experimental/python` ([databricks#3540](databricks#3540))