Clarify subgroup permissions

See merge request gitlab-org/gitlab-ce!21769

CE upstream - 2018-09-28 09:21 UTC

See merge request gitlab-org/gitlab-ee!7585

Documentation: Fix broken link of feature proposal template in issue_workflow.md

See merge request gitlab-org/gitlab-ce!21894

Fix karma out-of-memory failures

See merge request gitlab-org/gitlab-ce!21912

Rename block scope local variable in table pagination spec

See merge request gitlab-org/gitlab-ce!21969

Change class .label to .badge for 2FA label in group members view

See merge request gitlab-org/gitlab-ce!21759

CE upstream - 2018-09-27 21:21 UTC

Closes #1801, gitlab-ce#50181, and gitlab-ce#51925

See merge request gitlab-org/gitlab-ee!7570

Fix html rendering in api settings documentation

Closes #7784

See merge request gitlab-org/gitlab-ee!7569

Docs: improve Issue Boards overview (to deprecate feature page)

Closes #50503

See merge request gitlab-org/gitlab-ce!21905

Docs: improve Issue Boards overview (to deprecate feature page)

See merge request gitlab-org/gitlab-ee!7483

Resolve "Geo: Does not mark repositories as missing on primary due to stale cache"

Closes #50211 and #44044

See merge request gitlab-org/gitlab-ce!21789

I attempted to refactor so that the caller of `wrap_method` passes in a
block, rather than a method name, but I was unsuccessful.

I kept getting the following error:

NoMethodError:
        undefined method `cache_method_output' for Repository:Class

If you can figure this out, then feel free to dry up these class methods
again without doing a send-within-a-send.

* Only truthy values are cached in Redis.
* All values are cached in RequestStore and in an instance variable.

- Artifacts left behind by a previous merge conflict resolution caused
  a table on the api settings page to render improperly. Removed the
  stray '=======' and fixed the problem.

Resolves: #7784
Signed-off-by: Robert Marshall <rmarshall@gitlab.com>

And reuse `Gitlab::Utils::StrongMemoize`.

There is a subtle behavior change required to reuse StrongMemoize in
this case. The early fallback check now occurs *before* reading the
memoized value instead of after.

I think this is fine since a memoized value should only exist if
`exists?` is also already memoized as `true`.

# Conflicts:
#	spec/services/quick_actions/interpret_service_spec.rb

[ci skip]

Extend reports to support security features

See merge request gitlab-org/gitlab-ce!21892

Fix blue, orange, and red colors

See merge request gitlab-org/gitlab-ce!21972

CE upstream - 2018-09-27 06:21 UTC

Closes gitlab-ce#51299

See merge request gitlab-org/gitlab-ee!7533

Fix a sanity framework test

See merge request gitlab-org/gitlab-ce!21956

Add right margin to board-card-number

See merge request gitlab-org/gitlab-ce!21939

Group weight icon and text on issue list and issue boards

Closes #7741

See merge request gitlab-org/gitlab-ee!7484

Adds an extra width to the responsive tables

See merge request gitlab-org/gitlab-ce!21928

QA: Resolve "Failure in qa/specs/features/browser_ui/1_manage/login/register_spec.rb"

See merge request gitlab-org/gitlab-ce!21963

Fix key name on dashboard component

See merge request gitlab-org/gitlab-ee!7558

Bump gitlab-workhorse to 6.1.1 to fix custom error handling

See merge request gitlab-org/gitlab-ce!21967