Enable frontend routing for package details

See merge request gitlab-org/gitlab!77762

Fully enable vue-rotuer in package registry to fully
Leverage Apollo cache and improve page performance

Changelog: changed

`Cloud Run` and `Cloud Storage` buttons in GoogleCloud :: Deployments tab

See merge request gitlab-org/gitlab!78095

Always autofill group slug

See merge request gitlab-org/gitlab!78152

Update MR head pipeline when pipeline event is published

See merge request gitlab-org/gitlab!78172

Use mock apollo for validation segment tests

See merge request gitlab-org/gitlab!78070

Replace window.confirm with GlModal for registration token reset

See merge request gitlab-org/gitlab!77883

This commit replaces window.confirm dialog with
a GlModal for the reset of the registration token.

Changelog: changed

Remove `deprecated` metric status

See merge request gitlab-org/gitlab!76069

Geo: Increase verification backfill job exclusive lease timeout

See merge request gitlab-org/gitlab!78240

Fix erroneous pending migrations command

See merge request gitlab-org/gitlab!78242

Update excon to v0.90.0

See merge request gitlab-org/gitlab!78147

Add active state to milestones fetch in filtered search

See merge request gitlab-org/gitlab!78153

Improve upgrade details

See merge request gitlab-org/gitlab!78208

Clean up expired agent activity events

See merge request gitlab-org/gitlab!75709

Add more implementation and version information for new LDAP setting

See merge request gitlab-org/gitlab!78145

Fix autoscroll to comments on issues

See merge request gitlab-org/gitlab!76340

Add ResizeObserver to set scroll position if a comment is
selected. Still has a bunch of unwanted side effects like
jumping when window is resized, or when sidebars are collapsed
on smaller screens.

Remove ResizeObserver on user interaction
This avoids a bunch of bad and terrible bugs breaking scrolling
when user does something like toggle sidebar or click on text
editor. That otherwise triggers a resize

Changelog: fixed

Remove cross-db modification exception in Clusters::Applications::Runner

See merge request gitlab-org/gitlab!78236

Rename task-runner to toolbox

See merge request gitlab-org/gitlab!78198

Replace legacy checkbox name in docs

See merge request gitlab-org/gitlab!78166

Add warning to docs about providing release notes when creating a tag

See merge request gitlab-org/gitlab!77354

Whenever an activity event is created for an agent, a worker is
scheduled to remove all but the most recent 200 events.

If the agent does not yet have 200 events, the worker still runs,
however it will not delete any records.

Events may be created frequently, so to prevent many workers running
unnecessarily they are scheduled with a delay of up to an hour, and
then de-duplicated until executed.

Changelog: added

Update Release documentation about storage

See merge request gitlab-org/gitlab!77911

Add empty deployment component

See merge request gitlab-org/gitlab!78003

Refactor dynamically called methods to be more explicit

See merge request gitlab-org/gitlab!78141

Update topic for CTRT

See merge request gitlab-org/gitlab!77927

Update quarterly reconciliation content

See merge request gitlab-org/gitlab!78159

Because I have observed staging take longer than a minute
to complete this perform. A lease timeout which is too
short is against the point of the exclusive lease. The
timeout should ideally be exactly as long as the longest
possible execution time, but it's ok if we make the
timeout a bit longer than needed. I don't know for sure
if 30 minutes is long enough, but it is known to be an
improvement.

Changelog: changed
EE: true

Docs: Update Runner registration heading to be more explicit

See merge request gitlab-org/gitlab!78121

Runner registration wasn't clear enough, so we reworded
to make it more obvious that you're preventing people
from registering runners

Document praefect's separate_database_metrics setting

See merge request gitlab-org/gitlab!78134

praefect['separate_database_metrics'] will put praefect metrics that run
queries on the database on a separate /db_metrics endpoint so that
customers can choose to configure a slower scrape interval. Also, with a
multiple Praefect setup, customers can configure only one Praefect node
to query the database.

Also add /db_metrics prometheus metrics to the gitaly/index.md.