The fault-tolerant/recovery mode of the Chevrotain-based parser isn't
needed anymore since the suggestion was made to be lexical-based in
PR #15688. Therefore, there is no more reason to carry the Chevrotain
baggage. The grammar construct for Chevrotain was converted into a
simple recursive-descen parser. This represents a reduction from over
1,100 LoC to a leaner parser of only 270 LoC.

Also, we remove the dependency to Chevrotain, reducing the front-end
bundle by approx 18 KB.

* Build in GH actions and push to Dockerhub head

* Update build-and-push-head-images.yml

Caching node_modules (see PR #19202) doesn't seem to _always_ work for
Percy CI workflow, there must be some non-deterministic behavior
somewhere.
Until we figure out that, skip node_modules caching for now to unblock
Percy CI runs.

* Update presto-kerberos-docker branch used by run-presto-kerberos-test

Point the Presto Kerberos integration test to a `hacked-dec-2021` branch of `metabase/presto-kerberos-docker`, which contains

* the test_data catalog, which was added under `add-test_data-catalog`
* an updated download location for the `forego` binary dependency, which we now host

Since Cypress is need, we need to cache ~/.cache/Cypress as well.

tl;dr AdoptOpenJDK is now Temurin (part of Eclipse Adoptium)

https://blog.adoptopenjdk.net/2021/08/goodbye-adoptopenjdk-hello-adoptium/

With the code being instrumented for code coverage, and another
additional time to analyze + send the code coverage to codecov.io,
the typical execution runs awfully close to the 10-min limit.
Hence, add some extra room.

* Add Kerberos support to Presto JDBC Driver

Adding Kerberos properties as DB details properties

Refactoring utility methods for dealing with additional options

Adding test for Kerb properties -> connection string

Adding GitHub action to run integration test

Start to move logic for capturing necessary files to new script, and calling that from mba run

Adjust settings in `metabase.test.data.presto-jdbc` to capture test parameters

* replace backport label with a more flexible backport command

* fix backport bot command

* upgrade to @percy/cypress 3.x

@percy/cypress 2.x uses deprecated @percy/agent which includes
an old version of puppeteer. The old version of puppeteer causes
`yarn` command to fail on M1 macOS environment.

Upgrade to @percy/cypress 3.x removes the dependency and thus
solves the issue.

The upgrade was done by running `npx @percy/migrate`.

* remove percy healthcheck task

This no longer exists in 3.x, see
https://github.com/percy/percy-cypress/blob/master/README.md#upgrading

* remove deprecated percy/exec-action

and moved `percy exec` to the correct command

* run 'yarn install' before running percy

* Tweak Codecov settings to account for front-end coverage
* Scope back-end coverage with the correct codecov flag
* Add enterprise/frontend

* Add the whitespace linter

* Fix whitespace linter errors [except for one file]

* Add a line that will intentionally break stuff to verify the linter is working.

* Ok, remove the line that caused the linter to fail.

* Use latest version of the linter

* Fix missing newline

Since it needs to grab the built JAR anyway, it's faster to obtain the
version.properties from the JAR.

This shaves off approx. 1 minute from Percy CI run.

It's redundant since the job explicitly installs the proper version in
the previous step, and also running that prematurely pulls a bunch of
JARs from Maven Central.

This should shave 20 seconds from the CI runs.

* Remove `project.clj` from Percy workflows

* Replace `project.clj` with `deps.edn` in `build-for-test` script

* Update reference to the `project.clj` in the comment