Add a repro for "Cannot create another filter value if it starts with the same value as an existing value" (#45837)

* Derive MySQLEnum from Text instead of TextLike

Fixes #44807.

This change is in accordance with https://dev.mysql.com/doc/refman/8.0/en/enum.html describing the ENUM types as a
kind of string.  All our string operations seem to work with MySQL ENUM fields.

fix(webapp/collections): Tweak the height of action menu items so they align well with icons (#45618)

* docs: extract requires a db with regex

* Update docs/questions/sharing/visualizations/table.md

Co-authored-by: Jeff Bruemmer <jeff.bruemmer@gmail.com>

---------

Co-authored-by: Jeff Bruemmer <jeff.bruemmer@gmail.com>

Include  cards from the root collection in the Query Validator Tool

[Fixes #45756]

* tests for drivers with spaces in their identifiers

* Add driver feature

* Fix tests

* Remove ambiguous alias for sqlite

* Try to fix bigquery and sqlite tests not seeing left-join support

* Add explicit support for left-join because of foreign-keys during test hacks

* Handle left-join support weirdness

* Fix test

* Remove flakiness from e2e repro

* add to flaky group anyway

* add to flaky group anyway

* fixup! add to flaky group anyway

* Fix falsy tabSelected/tabRenamed checks

* Update URL management (WIP)

* Fix tests failing because of whitespace changes

* Fix tests failing because of parameter order

* Use one hook for parameters and tab URL management

* Fix tests failing because of parameter order

* Fix test failing because of whitespace changes

* Revert

* Fix DashboardTabs tests

* Fix tests failing because of parameter order

* Fix test

* Fix test failing because of whitespace changes

* Remove unit tests that should be e2e tests now

* Fix tab navigation

* Fix string encoding/decoding

* Revert test changes

* Consistent question search params

* Update DashboardTabs tests

* Fix `SyncedParametersList`

* Fix e2e tests

* Fix nav usage

* Fix types

* Remove `useSyncedQueryString`

* Fix test

* Move slug utils to dashboard utils

* Fix nav

* Revert nav change

* Fix e2e tests

* Fix public/embedded dashboards, handle site url

* Add repro test

* Bring back and fix `PublicOrEmbeddedDashboardPage` tests

* Move `SyncedParametersList` to qb

* Do not render if there is no time to format

* Add a unit test

* Render time if 0

* Simplify the logic with explicit check for `undefined`

* Refactor EditoTabs
- get rid of options prop
- use named export

* Remove unused data-testid

* Remove redundant id and aria-labelledby

* Remove redundant props spread

* Bring back data-testid in use
- Thanks VSCode for randomly showing me search results or not...

* Fix navigation issue

* Add a repro for first issue

* Adjust EntityPickerModal root element so that it works with "be.visible" assertion in Cypress

* Add a repro for second issue

* Refactor tests
- Fix not enough history entries issue

* Change order of tests to start with simpler one

* Fix invalid request issue

* Add more assertions

* Format code

* customize popover z-index

* make popover z-index customizable

* make the popover z-index customizable

* add tests for popover z-index

* also specify z-index for mantine popover

* Reuse connection for test data loading & JDBC data loading overhaul

* Revert silly change

* Remove unneeded logic for recording which datasets are loaded or not-yet-loaded

* Ok so I guess we do need the code to detect duplicate dataset loading after all

* B I G  improvements

* Appease Kondo

* Test fixes 

* More test fixes 

* Fix dataset-already-loaded?-test

* Fix test

* Fix Presto JDBC and greatly improve test data load speed for it

* Postgres test fix 

* Fix compilation errors

* MySQL fixes 

* Fix syntax error

* MySQL fixes 

* Implement `dataset-already-loaded?` for Athena and cache the set of existing databases

* Fix Redshift

* Fix Presto JDBC

* Vertica test fix 

* Fix Oracle dataset-already-loaded?

* Test fixes and parallelize some API tests.

* Test fixes 

* Oracle test fix? 

* Another test fix 

* Stop putting random spaces in code bodies this is not JavaScript code

* with-temp-vals-in-db should merge in original value of Database settings

* The most test fixes 

* 3 or 4 test fixes 

* Add schemas around stuff to catch errors

* Appease Kondo

* Implement dataset-already-loaded? for vertica

* Implement dataset-already-loaded? for SQLite and Snowflake

* [WIP] Testing driver filters

* Only test returned values in CI

* Add :uuid-type for dbs that support that type (test oracle's support)

* Try again for oracle test, redshift does not support uuid-type

* Add tests, support where available

* Add empty/null tests

* Adjust test query for athena

* Reset logging

* Fix comments

* Add Athena uuid conversion

* Address PR review

* Add comments

* Address PR comments

* Use varchar for portability. Add inline uuid casting for athena

---------

Co-authored-by: Ngoc Khuat <qn.khuat@gmail.com>

* fix tab actions not working in firefox

* add e2e test

* update e2e tests that need awaitRequest: false

* ephemeral pr envs on label

* Field Filter Value Normalization in Embedding Endpoints

Fixes 27643

Since we use query-params to pass paramter values, we have to normalize the parameter map values.

This PR is a WIP, as I'm trying to find a way to do this hollistically/properly.

First commit here is mostly as a marker for where to start working on this properly, as I think the json parsing won't
catch all cases, nor should the normalization happen without some attempt to use a schema to properly validate, maybe
even coerce things.

* read strings if boolean or numbers, replace blank str with nil

Also update the common dashboard param values fn to use embedding params from the URL blob if the app db doesn't have
embedding params. This happens upon first preview where nothing will have been written to the Appdb yet.

* Add query-param normalization test, add related embedding test

* schema didn't work in this case after all.

Worth a shot, but don't want this to block the PR, so reverting to the previous schema

* Add comment pointing out the downsides/reasons for this solution

It's nto a perfect solution, but it's probably the best we can get at the moment, without adding a new filter
type (boolean) and/or improving the design of hte embedding endpoints and how filters/parameters work

When we run queries, we preprocess them, during which the sandboxing
middleware adds information that we use when checking permissions
later on in query processing.

When we checked data permissions before saving a card, we were checking
permissions on the raw query, without preprocessing. Without the data
provided by the sandboxing middleware, the permission check sometimes
failed (depending on the permissions for other tables in the DB).

Note the big hack here: preprocessing can fail, for lots of reasons.
This broke a *lot* of tests that were passing in invalid queries. If an
exception occurs during preprocessing, we catch and just use the
original query like we were before.

* Fix deserialization of mongo native query string

* Add test

* Don't return duplicate cards in the Query Validator Tool

[Fixes #45673]

* Sync Snowflake `VARIANT` type as :type/SnowflakeVariant`

* Test fix 

* install before running script

* freeze dat lockfile