These functions are implemented with hairy regular expressions, and
it's more user-friendly and future-proof to name those functions in MBQL
rather than baking the `regexextract` and regex into the user's query.

It lets us evolve the regexes in the future if we detect a bug, and it
improves the UX since the user sees a meaningful function instead of

    regexextract([My URL Column], "(?<=[@\.])(?!www\.)[^@\.]+(?=\.[^@\.]{1,3}\.[^@\.]+$|\.[^@\.]+$)")

Also refactors the regexes somewhat so that they work for emails as well as URLs, and there's always just one layer of `:regex-match-first`.
Previously this was separated into two steps: URL or email to host, and host to (sub)domain.

Part of the follow-up for Extract Column epic #38964.

Co-authored-by: Noah Moss <noahbmoss@gmail.com>
Co-authored-by: Noah Moss <32746338+noahmoss@users.noreply.github.com>
Co-authored-by: Nick Fitzpatrick <nick@metabase.com>
Co-authored-by: John Swanson <john.swanson@metabase.com>
Co-authored-by: Sloan Sparger <sloansparger@users.noreply.github.com>
Co-authored-by: Sloan Sparger <sloansparger@gmail.com>

When we actually run the job to cache models, we select "refreshable
models", which selects `PersistedInfo` where:

- the `state` is something refreshable, and

- the question marked for caching is actually a model, and

- the question hasn't been archived.

However, when we selected `PersistedInfo`s to display to the user, we
didn't apply any conditions. Instead, we included PersistedInfo where
the question was not a model, or where the question had been archived.

To prevent confusion, let's hide these.

Fixes https://github.com/metabase/metabase/issues/41318

The filters for _numeric_ PK and FK filters are slightly different than a general numeric column. That's fine for numeric keys, but with string keys that same set of numeric filters was being returned, and then the ones for strings were removed. That left only `Is` and `Is Not` for a string PK or FK.

This PR allows more flexibility in filtering by string PKs and FKs, by checking the base type of the column and using the right set of filters for the type of key.

Closes #40665.

* FeedbackModal for the Embedding Homepage

* add source and change to correct endpoint

* fix tests, adding await and fixing endpoint

* use toEqual to also check for the source

* addresses PR feedback

* button becomes Send when email is written

* make 'comments' optional, trim inputs and refactor tests

* product-feedback endpoint (#40974)

* mvp of the proxy endpoint for the product feedback

* cleanup and fixes

* email :string -> ms/NonBlankString

* revert my solution

* Dan diff file

* makes fields other than source optional

* don't use default value in dev, to not spam the prod endpoint

* change from setting to function for product-feedback-url, as it's not meant to be changed by end users

* changed endpoint as per https://metaboat.slack.com/archives/C010ZSXQY87/p1712918951362979?thread_ts=1704910781.153589&cid=C010ZSXQY87



* fix wrong dismiss reason sent for 'I'm not interested right now'

* Update frontend/src/metabase/home/components/EmbedHomepage/EmbedHomepage.tsx

Co-authored-by: Mahatthana (Kelvin) Nomsawadi <me@bboykelvin.dev>

* renamed field: feedback -> comment

---------

Co-authored-by: Mahatthana (Kelvin) Nomsawadi <me@bboykelvin.dev>

This reverts commit e0e4ade0.

* Cancel statement on completion of execute-reducible-query

* Address review remarks

Co-authored-by: Noah Moss <noahbmoss@gmail.com>
Co-authored-by: Noah Moss <32746338+noahmoss@users.noreply.github.com>
Co-authored-by: Nick Fitzpatrick <nick@metabase.com>
Co-authored-by: John Swanson <john.swanson@metabase.com>
Co-authored-by: Sloan Sparger <sloansparger@users.noreply.github.com>
Co-authored-by: Sloan Sparger <sloansparger@gmail.com>

Not all drivers support regular expressions. Don't return the
Column Extractions for URL -> (sub)domain and email -> domain unless the
database can support the regular expression matches.

Part of the follow-up for Extract Column epic #38964.

This improves our Malli performance substantially! Many of our code paths were calling vanilla `malli.core/validate` or `malli.core/explain`; these redo the (possibly expensive) compilation of the schema into a validator or explainer for every call.

We should use the caching versions in `metabase.util.malli.registry` everywhere, and our runtime performance will be much improved.

However when I started using these cached versions, I found that the memory use was growing out of control. Eventually I tracked this down to `:fn` schemas. Functions are only comparable by pointer equality, so they make poor cache keys. `(fn ...)` or `comp` calls in a schema on a `mu/defn` function **get re-created for every call of the `mu/defn`'d function**! That's a big time sink recompiling the schemas if we're *not* caching, and a huge memory sink if we *are* caching!

This PR pulls every such schema I could find out into a `def` so it uses the same closure and is cacheable. I'd like to automate that in some part of the Malli pipeline, or maybe a linter rule, but I haven't found a good way to do it yet.

Part of #39946.

The drill-thru is mostly a stub to put an entry in the context menu. Most of the logic for this is built on the FE using the existing functions for consuming and creating expressions.

See [thread](https://metaboat.slack.com/archives/C06P22KS4JH/p1712264174056569) for a lot more context.

This also adds `docs/developer-guide/mbql-library-changelog.md` and documents adding this function to the library as the first change in the API. I'll follow up in another PR with Github automation to expect the changelog to be updated whenever `src/metabase/lib/js.cljs` changes.

Fixes #39978.

* Keep Filter Values for enabled Parameters in Embedding

Fixes: #37914

Previously, we removed locked parameters entirely. This is to prevent leaking potentially sensitive values.

However, for the situation where the same field backs 1 locked or disabled paramter and 1 enabled parameter, we do
still want to send the paramater values, because the enabled parameter implies that the values are permissible to see
in the embed.

So, this change will still remove parameters based on their 'disabled/locked/enabled' status, but will NOT remove the
linked field ids if they're also being used by 'enabled' parameters.

This results in the backend correctly sending necessary parameter values to the embed, where the frontend can then
render the appropriate UI instead of falling back to just text filters.

* test the case where a locked and enabled param share same field

* Address feedback. Added comment to explain classify fn

* Exclude implicitly joinable columns from visible-columns in fk-columns-to

* Add test

* Update description

[P1 BUG FIX] Cumulative aggregations with multiple breakouts should reset counts when values of breakouts other than change (#41340)

* Fix add-alias-info matching for nominal field literal refs

* Minor cleanup

* Minimize diff

* A bit more cleanup

* Cumulative aggregations w/ multiple breakouts should reset counts (#2862)

* Update dox

* Do the nesting FIRST before adding alias info.

* Remove unused

* Fix add-alias-info matching for nominal field literal refs

* Minor cleanup

* Minimize diff

* A bit more cleanup

* Convert `auto-bucket-datetimes` middleware to MLv2

* New Kondo rule time.

* Test & lint fixes 



* Fix typos

* Apply suggestions from code review

Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

---------

Co-authored-by: metamben <103100869+metamben@users.noreply.github.com>

* collection and db breadcrumbs

* compact breadcrumbs

* adding navigation keybindings

* recent_views and search changes to always include collection or db name

* removing breadcrumbs, display parent name

* removing unneeded components

* e2e and FE unit tests

* waitFor

* e2e tests for keybindings

* help

* be tests passing

* PR feedback and type adjustments

* unit test adjustment

* Test created_at and schema fields
* Coerce nil schema to empty string