backported "fix(sdk): migrate to custom redux context to allow using the sdk on host apps that use redux" (#51414)

* step1: add MetabaseReduxProvider and make our hooks + connect fn use it

step1: add MetabaseReduxProvider and make our hooks + connect fn use it

* step 2: codemod

* step 3: manual fixes

* step 4: e2e test for sdk

---------

Co-authored-by: Nicolò Pretto <info@npretto.com>

* lychee to lint doc links (#51422)

Co-authored-by: Ryan Laurie <iethree@gmail.com>

* tests(jest): add ES module is-absolute-url to the transform ignore (#51438)

---------

Co-authored-by: Jeff Bruemmer <jeff.bruemmer@gmail.com>
Co-authored-by: Ryan Laurie <iethree@gmail.com>
Co-authored-by: Uladzimir Havenchyk <125459446+uladzimirdev@users.noreply.github.com>

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

Co-authored-by: Chris Truter <chris@metabase.com>

* revert me - run this workflow against the release-x.52.x branch

* Revert "revert me - run this workflow against the release-x.52.x branch"

This reverts commit 09026adc.

* re-enable cross-version tests

Co-authored-by: Chris Truter <chris@metabase.com>

Co-authored-by: Alexander Polyankin <alexander.polyankin@metabase.com>

Co-authored-by: Phoomparin Mano <poom@metabase.com>

Co-authored-by: Demetrio <53023423+ddowbnac@users.noreply.github.com>

* temporarily disable e2e cross-version tests

* Revert "temporarily disable e2e cross-version tests"

This reverts commit b5c60946.

* skip until workflow is fixed

* Fix Embedding SDK cross-version e2e tests for release 51 branch (#51171)

* bogus change to trick the ci - to revert

* add the stub back

* fix the stub name

* fixme - hard code sdk_version for testing the workflow

* fixme - hard code build requirements for testing the workflow

* check resolved package path

* fixme - check why require.resolve is erroring

* fixme - add console log to get resolved package path

* fixme - additional ci debugging logs

* fixme - try limiting module search path to node_modules

* fixme - try multiple strategies for resolving modules

* fixme - throw a script error if resolution fails

* remove unused event.json

* fix webpack component script

* fixme - comment out pull_request event block

* return empty alias path if resolution failed

* temporarily skip create question tests

* fixme - debug create question tests failing

* fixme - debug create question tests failing

* fixme - revert sparse checkout version pinning

* fixme - change hard-coded sdk version

* revert fixme debugging changes

* cleanup workflow changes

* cleanup fixme changes to trigger ci

---------

Co-authored-by: Denis Berezin <denis.berezin@metabase.com>

* Fix unexpected rounding in csv & json exports when column multiplier is non integral

* Fixing table unit test

* Adding WIP test

* Adding test case for validating new scaling logic is applied appropriately in exports

* Reverting id-or-name change and adjusting new scaling unit test

* Updating formatting to fix cljfmt in CI

* removing debug def

* Preserving integral behavior if scale is nil

Co-authored-by: Tyler Plude <tsplude@gmail.com>

Co-authored-by: Nicolò Pretto <info@npretto.com>

Co-authored-by: Phoomparin Mano <poom@metabase.com>

Co-authored-by: Uladzimir Havenchyk <125459446+uladzimirdev@users.noreply.github.com>

Previously, cumulative aggregators were being done in clojure after the mongo query was finished.  However, this
didn't work on multi-step queries where the cumulative aggregator wasn't the last step (for example, do a cumulative
sum and then filter the result).  Implementing cumulative operators directly in mongo fixes this issue.

Co-authored-by: William <william@metabase.com>

* wip nextjs compat docs

* explain how to do manual wrapping

* doocs

* better explanation of the index file

* Apply suggestions from code review



* Update docs/embedding/sdk/next-js.md



---------

Co-authored-by: Nicolò Pretto <info@npretto.com>
Co-authored-by: Jeff Bruemmer <jeff.bruemmer@gmail.com>
Co-authored-by: Phoomparin Mano <poom@metabase.com>

Co-authored-by: Nicolò Pretto <info@npretto.com>

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

Co-authored-by: Phoomparin Mano <poom@metabase.com>
Co-authored-by: Mahatthana (Kelvin) Nomsawadi <me@bboykelvin.dev>
Co-authored-by: Phoomparin Mano <phoomparin@gmail.com>

Co-authored-by: Oisin Coveney <oisin@metabase.com>

Co-authored-by: Oisin Coveney <oisin@metabase.com>

Co-authored-by: Oisin Coveney <oisin@metabase.com>

Co-authored-by: Oisin Coveney <oisin@metabase.com>

Co-authored-by: Oleksandr Yakushev <alex@bytopia.org>

Co-authored-by: Oleksandr Yakushev <alex@bytopia.org>

* Add missing -test suffix to zoom-in-timeseries test name

* Add missing -test suffix to underlying-records test name

* Fix distribution drill thru for multi-stage queries

Ensure the distribution drill thru is not returned by available-drill-thrus when a user clicks on an aggregation or
breakout column, even if there are additional query stages after the last aggregation / breakout stage.

Co-authored-by: appleby <86076+appleby@users.noreply.github.com>

Co-authored-by: Aleksandr Lesnenko <alxnddr@users.noreply.github.com>

Don't compute a different hash key if they only differ on the order of values in a parameter (#51186) (#51306)

Don't compute a different hash key if they only differ on the order of values in a parameter

Co-authored-by: Nathan Voxland <nathan@metabase.com>

Some funcs were added to lib.drill-thru.common for the recent fix to the zoom-in.timeseries drill. Update the
underlying-records drill to use the common lib code rather than duplicating it.

Co-authored-by: appleby <86076+appleby@users.noreply.github.com>

Co-authored-by: Ryan Laurie <30528226+iethree@users.noreply.github.com>

It looks like snowflake introduced some very noisy logging in their
code.

https://github.com/snowflakedb/snowflake-jdbc/commit/9b7c9eb3e02c93cf7788c9866e1a483fe978e1be#diff-7859cac7243b3b9668df0770c96e264a1be2be06eefbbf7bba63904104028495L743



This should reduce the noise and allow our code to handle and log the
exceptions.

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

We changed the time_history icon in this context to a revert icon in #46050 but never updated this test. This PR fixes that.

Co-authored-by: Raphael Krut-Landau <raphael.kl@gmail.com>

* save only grid on fixed width dashboards

* split export pdf into pages

* trim dashboard export pdf into pages

* specs

* add padding to pdf pages

* dark mode

Co-authored-by: Aleksandr Lesnenko <alxnddr@users.noreply.github.com>

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

Co-authored-by: Oisin Coveney <oisin@metabase.com>

* Fix #51020

* Add a repro for #51020

* Improve test name

* Update comment and naming

* Improve diff

* Add repro for when primary key is not called 'id'

* Update repro

* Drop prefetchesInitialValues & getValuesForUpdate

It's unclear what the real intention behind these two was, but it's clear that it made two differences:
(1) It decreased the probability of overwriting someone else's changes (if someone else saved them after we opened the form)
(2) It decreases the payload sent over the wire

(1) We can ignore this, as this problem is prevalent everywhere in our app
(2) This is negligible

Dropping these is simpler than enhancing getValuesForUpdate with extra logic for primary keys (and who knows what else)

* Break beforeEach into 2

* Remove no longer valid test

Co-authored-by: Kamil Mielnik <kamil@kamilmielnik.com>

should be reverted but we want a 52 release

context:
https://github.com/metabase/metabase/actions/runs/12307551790/job/34351648561