* Add upterm session

* Add sleep to avoid race condition

* Remove upterm session

* Temporarily remove flaky test

* Remove sleep

* refactor custom `TabList` from `ModelDetailPage` into a core component (#29669)

commit-id:e6ea2eb9

* add menu to `TabButton` (#29697)

* add `disabled` prop to `TabButton` (#29772)

* create renameable `TabButton` variant (#29801)

* add `showMenu` prop and fix bg color

* Rename stage field to fields

* Convert join condition to conditions

* Do legacy to mbql conversion in lib.util/pipeline

* Fixing tests

* Do not convert questions twice

* Fix linter

* Fix tests

* Fix mongo tests that were building incorrect legacy queries

* Fix mongo properly

* Fix #29902

* Test fixes & PR feedback

* Re-export typesafe redux hooks

* rename `redux.unit.spec.js`

* add support for `dispatch.action` syntax

* use `createMockUser`

* replace author name in comment with PR number

* remove custom hooks in favor of using `extend-redux`

* revert changes to `renderWithProviders`

Added code to infer some basic visualizations once sql is inferred. Includes tests.

This reverts commit 1f402637.

* MLv2: Add sample-dataset test metadata

* Fixups

* More fixes 

* Add safe DDL commands to the whitelist for actions

* Run 'GRANT ALTER ANY SCHEMA TO GUEST;' on sample database

* Grant permissions for test databases and add test for sample database

* Fix test

* Fix test

* Add test for adding/removing columns

Limiting viewable LLM modes to the latest version of each generation of GPT.

* MLv2 source-alias and desired-alias

* Add tests for JVM metadata provider

* De-duplicate JVM versions

* Remove unused

* Appease Kondo

* Address PR feedback

* Test fixes 

* Fix padding logic.

* Extra test cases

* Add test for saved question / model as a source

* Fix Model metadata

---------

Co-authored-by: Cam Saul <github@camsaul.com>

* orderable-columns should not return columns that are already in order-bys

* Add test for table + join

* More tests 

* MLv2 display-info method

* Update JS wrappers

* Appease TypeScript

* Remove unused import

* Use hierarchy for new multimethod

* Address PR feedback

Adding a more generic SQL inner query along with expected tests for supported dialects.

* Fixes backfill-ids/has-entity-id?
* Adds hydration for Action :model
* Adds test for Action :model hydration

* hide ask metabot menu items when the model db is not supported

* check for metabot support in other places

Modifying the metabot `supported?` query to be more generic. The intent of the query is to just see if the db supports a basic SQL select from a CTE.

* Add direction arg to orderByClause

* Update order_by.ts

* Fix TypeScript error

* Update order_by.ts

---------

Co-authored-by: Anton Kulyk <kuliks.anton@gmail.com>

* update and test number display for gauge viz

* update and test numeric display for progress viz

* display null in scalar viz

* Rework repro for 16382

* Merge multiple "connection settings" tests together

* Merge multiple "scheduling settings" tests together

* Merge tiny test with the main body

* Remove redundant tests

* Rework the sample database action sidebar

* Handle database exceptions in one spec

* Move stray database test to exceptions spec

* Move Postgres SSL test to external

* Group together Google service account JSON tests

* Custom caching

* Move remaining pieces of add to external

* Rename spec to `add-new-database`

* Move repro for 20471 to exceptions

* Add segments and metrics to the database we want to delete

* Add `visitDatabase` helper

* MLv2 Multimethod hierarchies

* Remove test

Added both FE and BE tooling for integration with LLMs, called "metabot". FE adds prompts for querying models and databases. BE takes these prompts, calls a LLM, parses the response, and returns the result.