For each pipeline, the controller will call `Pipeline#latest?` to
determine if the pipeline's ref is the latest for that branch.
Since it's likely that the same branches are being used in each
pipeline, we can reduce Gitaly overhead by caching the results
of the FindCommit call.

For a given merge request, it's quite common to see duplicate FindCommit
Gitaly requests because the Gitaly CommitService caches the request by
the commit SHA, not by the ref name. However, most of the duplicate
requests use the ref name, so the cache is never actually used in
practice. This leads to unnecessary requests that slow performance.

This commit allows certain callers to bypass the ref name to
OID conversion in the cache. We don't do this by default because it's
possible the tip of the branch changes during the commit, which
would cause the caller to get stale data.

This commit also forces the Ci::Pipeline to use the full ref name
so that caching can work for merge requests.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/57083

In production, we see high error rates due to clients attempting to use
the dumb Git HTTP protocol with HEAD /foo/bar.git/info/refs
endpoint. This isn't supported and causes Error 500s because Workhorse
doesn't send along its secret because it's not proxying this request.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/54579

When executing quick actions, this limits the `commands_changes`
response to only those used by the frontend

Move this to EE-only. It isn't useful here since there's only one board

Move common functionality to BoardsActions concern

Update spec

Returns all diff lines for frontend if full is true.
Turn UnfoldForm into presenter, and move controller logic to presenter.

- This is to avoid colision with EE ClusterUpdateAppWorker

- Creates new route
- Creates new controller action
- Creates call stack:
  Clusterss::ApplciationsController calls -->
  Clusters::Applications::UpdateService calls -->
  Clusters::Applications::ScheduleUpdateService calls -->
  ClusterUpdateAppWorker calls -->
  Clusters::Applications::PatchService -->
  ClusterWaitForAppInstallationWorker

DRY req params

Adds gcp_cluster:cluster_update_app queue

Schedule_update_service is uneeded

Extract common logic to a parent class (UpdateService will need it)

Introduce new UpdateService

Fix rescue class namespace

Fix RuboCop offenses

Adds BaseService for create and update services

Remove request_handler code duplication

Fixes update command

Move update_command to ApplicationCore so all apps can use it

Adds tests for Knative update_command

Adds specs for PatchService

Raise error if update receives an unistalled app

Adds update_service spec

Fix RuboCop offense

Use subject in favor of go

Adds update endpoint specs for project namespace

Adds update endpoint specs for group namespace

- Move some specs into contexts
- Let get_slugs method take a parameter and return a specific slug.
- Add rescues when using Addressable::URI.

It's needed to show shorter names in environments "folders"

The repository charts page used to detect the repository language for
each request that was made to the page. Given the detection is an
expensive operation and the same data is stored in the database the
database is now serving the request.

The same goes for an API endpoint that serves the languages.

When a repository is empty or non-existent the languages will always be
empty. And the language detection RPC isn't requested.

Closes: https://gitlab.com/gitlab-org/gitlab-ce/issues/47390

Prevent unauthorized users having access to milestone titles
through autocomplete endpoint.

Move remove_pages permission to maintainer
Fix before_action in pages controller to check `remove_pages`
permission
Add specs

Removes unneeded params from permitted list

This also fixes commenting on commits with a hash starting with a large
number

These specs were flaky and only passing after a retry due to how
rspec-retry works.

1. The test with paths that don't exist was returning 200 on the first
  try, then 404 on the second, not because the paths don't exist, but
  because the MR IID didn't change, thus the MR couldn't be found.
  I decided to remove the test entirely since we don't seem to return
  404 for paths that don't exist.
2. The test with a user that cannot view the merge request was failing
  the first time because the project owner wasn't removed with
  `project.team.truncate`.
  Changing the `let(:user)` to `create(:user)` and calling
  `project.add_maintainer(user)` in the `before` block fix the test.
Signed-off-by: NRémy Coutable <remy@rymai.me>

Add changelog

Rename word to query

User hash for limit

Do not allow control limit

Rename pluck names and add more specs

Users downloading non-ASCII attachments would see garbled characters.
When used with object storage, AWS S3 would return an InvalidArgument
error: Header value cannot be represented using ISO-8859-1.

Per RFC 5987 and RFC 6266, Content-Disposition should be encoded
properly. This commit takes the Rails 6 implementation of
ActiveSuppport::Http::ContentDisposition
(https://github.com/rails/rails/pull/33829) and ports it here.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/47673

Removes special error message when creating new issues

Re-enables and autocorrects all instances of the
Style/MethodCallWithoutArgsParentheses rule

In order to let users' sorting preferences transfer between devices, we
save the preference for issues and MRs (one preference for issues, one
for MRs) in the backend inside the UserPreference object

This includes a set of APIs to manipulate container registry.
This includes also an ability to delete tags based on requested
criteria, like keep-last-n, matching-name, older-than.

Adds to search ILIKE search for milestones title in:
  - Milestones dashboard
  - Group milestones page
  - Project milestones page

Signed-off-by: NRémy Coutable <remy@rymai.me>

This is to match `git branch -D <branchname>`

To support environment folders in the UI on the Environments List
page, the environments index route previously returned one environment
per folder, excluding those other than the latest deploy. However, the
environtments dropdown on the metrics dashboard requires that any
environment be selectable.

To accommodate both use cases, support an optional 'nested' parameter
in the index route to return either a flat, complete response or a
nested response based on the use case in question. The new default
response structure is the flat response.

We introduced the feature flag `error_tracking` to test the Sentry MVC
safely.

After the successful test, we remove it again.