* remove internal user on rollback + test

* fix naming + add newline

* fix typo

* rollback change to migration changeset ID

- since it would alter the checksum for it

This reverts commit adfda064 now that #35621 has fixed the underlying issue.

* respect MB_PLUGINS_DIR on serialization load

* respect MB_PLUGINS_DIR on serialization load

* refactor use-list-select hook

- toggleAll(items) felt unintuitive as some items could be selected while simultaneously unselecting others
  - switched to selectOnlyTheseItems(items) to be more explicit on what the function is doing
- updated consumers of useListSelect to use the new API

* convert to functional component

* convert HoC ListSelect to hook useListSelect

* convert redux connects to useSelectors

* refactor ArchiveApp and improve types

* hide read only items in the archive

* verify desired behavior

* refactor beforeEach

* remove unused eslint disable line

* refactor item can_write accessor logic

* Add :can_write key into the search data response. uses `mi/can-write?`

* update archive app for archived search changes

* refactor filtering of list

* replace hard-coded collection ids in spec

* move specific helper outside of generic helpers file

* remove admin gets unfiltered data

* BE sends only can_write true results when search with archived?=true

This change modifies the `check-permissions-for-model` method to check for write access if the SearchContext has
archived true.

This means that the only results sent to the frontend are those results that the user can take actions on, namely,
unarchiving them.

* reomve unused require

* update FE to account for BE changes

---------

Co-authored-by: adam-james <21064735+adam-james-v@users.noreply.github.com>

* Audit db sync cannot happen

- disable all job scheduling for audit db
- make syncing/scanning auditdb a noop in prod (throws otherwise)

* better name for job-data -> jobs-to-create

* test no scheduling and no running audit db sync

* sort ns

* fix trivial check in test

* use metabase.task/scheduler-info intead qrtz table

* sort ns

* tweak wording in config file docs

* Update docs/configuring-metabase/config-file.md

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

* add period

---------

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

* command page

* formatting

* command updates

The QP returns "large int" IDs like FKs as string, since JS doubles only
guarantee 51 bits of mantissa and a 64-bit ID might get rounded.
However NULL IDs were getting returned as `""` empty strings.

This fixes the QP middleware to return `nil` (JS `null`).

In addition, the drill-thru logic in the FE and MLv2 was trying to
filter incorrectly with a NULL value. This PR prevents the FK filter
drill from returning anything in that case.

I filed #35561 to track the fact that would be useful to drill thru to
all rows with a NULL for some FK, but it's low priority.

Fixes #13957.

Follow up after #34892 

* Include E2E workflow in file-path triggers
* Adjust the glob for "actions" directories

* avoid calling fetchCardData in undoRemoveCardFromDashboard for virtual dashcards

* test: adds basic e2e test for text card

* Fix flake for repro 20393

* Fix flake for repro 22524

[ci skip]

[ci skip]

This reverts commit d140d676.

Co-authored-by: Case Nelson <case@metabase.com>

It appears the code fails when run in parallel due to the `metabot-settings/enum-cardinality-threshold` value.

To demonstrate, I ran:

```clojure
(frequencies
  (pmap
  (fn [_]
    (clojure.test/run-test inner-query-name-collisions-test))
  (range 1000)))
```

and got:

```
{{:test 1, :pass 12, :fail 0, :error 0, :type :summary} 414, {:test 1, :pass 11, :fail 1, :error 0, :type :summary} 586}
```

Running with a simple sequential map produced no errors.

Wrapping the second test in `tu/with-temporary-setting-values [metabot-settings/enum-cardinality-threshold 10]`
reduced the error count dramatically (in the teens for 1000 samples), but still wasn't bulletproof.
I'm guessing the two `testing` directives in the test are run in parallel and the setting is not thread safe.

By adding the explicit threshold to the test and breaking it out into a standalone test the
error count drops dramatically.

```
(frequencies
  (pmap
    (fn [_]
      (clojure.test/run-test inner-query-name-collisions-with-joins-test))
    (range 1000)))
;=> {{:test 1, :pass 6, :fail 0, :error 0, :type :summary} 998, {:test 1, :pass 5, :fail 1, :error 0, :type :summary} 2}
```

I'm still not sure why this fails (only twice in 1000 runs), but apparently something about
this operation is not thread safe. However, given the additional isolation added by this PR
we go from a 60ish percent failure rate down to a 0.2% failure rate locally.

Hopefully this is adequate enough to prevent any substantial level of future flakes.

* Audit log CRUD permission failures

When we run `metabase.api.common/*-check` functions like `update-check` or `create-check`, publish an event when the
check fails, along with enough context that, on the other side, the audit log handler can record the relevant event.

Note that we won't fire events for a `read-check` failure. Eventually we will and these events will be handled by the "read log" instead of the audit log.

If we throw an event that isn't derived from `:metabase/event` it will throw an exception. That `derive` no longer belongs in the
audit log code, and until we have the view log there's not really a logical place to put it. For now, I just commented
out the `publish-event!` call, with the assumption that the view log is coming very soon. Otherwise we can just delete
it and add it back in when it's needed.

Some tangentially related required changes here to make this work include:

* allow `audit-log/model-name` to work with RootCollections

`(t2/model root-collection)` returns a Class rather than a keyword - `name` doesn't work here, so we'll use
`getSimpleName` instead.

* add a default model-details clause for dashboards

If the event-type is not one that we've explicitly delineated, just return an empty map for the model details.

* Migration to fit longer model names in `activity.model`

The `NativeQuerySnippet` or `PermissionsGroupMembership` models, for example, were too long to fit in `VARCHAR(16)`.
I'm doubling the length of this column to fit the longest model I see, with a little extra room.

Follow up after #34209

Co-authored-by: bryan <bryan.maass@gmail.com>