* Split from #40146: only the filter changes

* Don't include format-rows changes here.

* Test fixes 

* Test fix 

* Misc test fixes 

* Test fix 

* Tweak SQL Server fix

They were showing "break out by" and quick filters, when they should be
showing "break out by" and "See these Orders".

These are the `pivot` and `underlying-records` drills, internally.

Fixes #40174.

* Sort official collections first in API endpoints

For both `/api/collection/tree` and `/api/collection/items`, sort
collections with official collections coming first.


Co-authored-by: Noah Moss <32746338+noahmoss@users.noreply.github.com>

Fixes #40383

If a native query parameter is set up with a string widget type that
supports a `case-sensitive` option, and that parameter is tied to a
dashboard filter that did not support that option, the option was still
leaking into the filter as an option where options were not expected.

For instance the native query parameter used `string/contains` and the
dashboard used `string/=`, the QP would try to build a filter like `[:=
field value {:case-sensitive false}]` which is an illegal filter.

Include postgresql tables that are visible through column grants instead of full table grants for simple questions (#40034)

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

* Funnel Chart Static-Viz should handle row data more robustly

The row data coming into the funnel chart render method can evidently have a few slightly different shapes, and may
need to be 'coordinated' with the :funnel.rows key from viz-settings (names and order are specified there).

This is a WIP to make the method more robust to different row data shapes. For example:

[[1 100] [2 200]] should work as well as [["A" 100] ["B" 200]] and correctly handle the viz-settings

WIP because I need to add a test or two, try to get a handle on the exact schema that is allowed for 'row shape' and
the funnel.row viz key, and see if there are failing cases that I haven't considered yet.

* Funnel rows should now work more effectively with possible raw-rows

* Add a test that checks the success of the render and correct order

* Update src/metabase/pulse/render/body.clj

Co-authored-by: Chris Truter <crisptrutski@users.noreply.github.com>

* Update src/metabase/pulse/render/body.clj

Co-authored-by: Chris Truter <crisptrutski@users.noreply.github.com>

* Addressing feedback from the review

 - renamed build-funnel-rows to funnel-rows
 - used 'funnel-viz' as the binding for the {:key "asdf" :name "asdf" :enabled true} maps from the viz settings
 - added docstring to the function to try clarify its 2 branches (keys vs indices on the rows)

---------

Co-authored-by: Chris Truter <crisptrutski@users.noreply.github.com>

* Trend Chart Viz Should Use Proper Card execute in Rendering

WIP

Fixes: #39854

The trend chart follows a code path shared with multi-series static-viz. This means there was an implicit expectation
that a dashcard is associated with the card at all times. This isn't true for alerts, so the code is fixed to remove
the failure.

* Add a test

* add missed :breakout that actually made the test work

oops

* Adding comments to try clarify what's going on

* /api/collection/:id/items gets `here` and `below`

During postprocessing of `collection` children retrieved by
`/api/collection/:id/items`, annotate each of the children with `below`
and `here` keys representing the presence of questions, models, or
collections at this level of the hierarchy or below it.

Co-authored-by: Noah Moss <32746338+noahmoss@users.noreply.github.com>

Fixes: #40306

Our datetime formatter relies on the `java-time.api`, for which there are many different, sometimes confusing, formatter patterns: https://docs.oracle.com/javase/8/docs/api/java/time/format/DateTimeFormatterBuilder.html#appendPattern-java.lang.String-

In this case, 'YYYY' is a week-of-year style year, which calculates which week a date falls into before returning the
year. Sometimes days near the start/end of a year will fall into a week in the wrong year.

For example, apparently 2023-12-31 falls into the 1st week of 2024, which probably not the year you'd expect to
see. What we probably do want is 'yyyy' which calculates what day of the year the date is and then returns the year
based off of that instead of the week number.

For an explanation, you can check out this SO answer: https://stackoverflow.com/a/46395342 provides an explanation.

* Wow

* Test fix 

* Fixes

* Actions should use strings for column names (fix :update-row and :create-row normalization)

* MLv2 schema should check against keys for the other query type

* Ok, have I fixed things?

* More fixes 

* Fix indentation

* Another round of test fixes. 

* Hopefully the last few test fixes 

* We need to test normalization for queries that have keyword keys as well.

* Fix Cljs i18n namespaces

* Sort namespaces

* Only test against H2

* Rename `metabase.mbql` to `metabase.legacy-mbql`

* Fix Kondo warnings

* Test fixes 

* Register MBQL clause schemas and test fixes 

* Test fixes and PR feedback

* Test fix

* Remove the normalization tests

* Test fixes 

* Fix kondo

* Fix import

* Another fix 

* Merge

* FIXES

* Add another missing REQUIRE

* E2E support for CRUD + running pMBQL queries; new MLv2 normalization

* More work on improved normalization

* Update schema references

* Test fixes 

* Move the MBQL parameters schema

* Performance improvements

* Improved pivot code etc.

* Clean namespaces

* Fixes 

* Wow

* Test fix 

* Fixes

* Actions should use strings for column names (fix :update-row and :create-row normalization)

* MLv2 schema should check against keys for the other query type

* Ok, have I fixed things?

* More fixes 

* Fix indentation

* Another round of test fixes. 

* Split pivot stuff off

* Hopefully the last few test fixes 

* We need to test normalization for queries that have keyword keys as well.

* Fix Cljs i18n namespaces

* Sort namespaces

* Only test against H2

* Test fixes 

* Register MBQL clause schemas and test fixes 

* Test fixes and PR feedback

* Test fix

* Remove the normalization tests

* Test fixes 

* Fix kondo

* Test fix

* Another fix

* Test fix 

* decrypt and read values from airgap token

- Adds notion of "AirgapToken"
- Adds max-users and company optional fields to `TokenStatus`
- Adds branch to `fetch-token-status*` for handling airgapped tokens

* Add support for handling airgap tokens

- validate airgap token valid-thru date

* enforce user creation limits using airgap token

* fix airgap token reads + fill in the token data

* in oss mode don't try to read the airgap_ token

* use airgap-token? helper

* tighten the user-creation logic

- make the check correct
- account for archived users

Co-authored-by: John Swanson <john.swanson@metabase.com>

* check airgap user count

setting premium-embedding-token
app startup

* move decryption code into ee namespace

* add some tests for token decryption

* fix tests and add tests

* move ee features into ee tests

* add typehint + warn on reflection

* add a check for missing public key resource

* respond to review

- add docstring to AirgapToken
- remove outdated comment
- fix off-by-1 error

* valid-now? takes a TokenStatus instead of a :map

* revert usage of mt/with-temporary-setting-values

---------

Co-authored-by: John Swanson <john.swanson@metabase.com>

* Rename :model/Metric to :model/LegacyMetric
* Rename ::lib.schema.metadata/metric tp ::lib.schema.metadata/legacy-metric
* Rename models.metric/Metric to models.metric/LegacyMetric
* Move /api/metric endpoints to /api/legacy-metric

This issue was previously punted, because the schema for aggregation
clauses allowed the tag to be `:any`. Now it's a short list of permitted
arithmetic expressions, but that list is too short and excludes things
like `:round`.

This change searches for an aggregation clause transitively in the
expression, and so allows any arithmetic expression.