Instead of flushing the behind/ahead counts for all branches upon every
push we now only flush the cache of branches that actually need to have
these statistics recalculated. There are now basically 2 scenarios and
their effects:

1. A user pushes a commit to the default branch, this results in the
   cache being flushed for all branches.
2. A user pushes to a non default branch, this results in _only_ the
   cache for that branch being flushed.

The existing code (Repository#expire_cache) remains backwards compatible
with the previous behaviour, the new behaviour is only applied when a
branch name is passed as an argument. This ensures that when for example
a project is deleted the cache for all branches is flushed.

This caches the output of the following methods:

* Repository#empty?
* Repository#has_visible_content?
* Repository#root_ref

The cache for Repository#has_visible_content? is flushed whenever a
commit is pushed to a new branch or an existing branch is removed.
The cache for Repository#root_ref is only flushed whenever a user
changes the default branch of a project. The cache for Repository#empty?
is never explicitly flushed as there's no need for it.

Fixes #3903

gitlab_git 8.1 adds the ability to count the amount of commits between
two references without having to allocate anything but regular
Rugged::Commit objects. This in turn speeds up the process of counting
the number of commits a branch is ahead/behind by about 3.5x.

Fixes #3339

This MR hides the 'Remove source branch' button when a new commit is added to the source branch

Currently any spam detected by Akismet by non-members via API will be logged
in a separate table in the admin page.

Closes #5612

Instead of running ClosingIssueExtractor for every commit in a merge
request we can gather all the commit messages (and the merge request
description), concatenate all this together and then run
ClosingIssueExtractor only once.

The result of this is that MergeRequest#closes_issues is now between
3.5x and 4x faster than the old setup. Using a merge request with 10
commits (each referencing a number of issues to close) this reduced the
call duration from around 200 milliseconds to around 50 milliseconds.

As a result of these changes the Jira related tests for
MergeRequest#closes_issues have been removed. These tests stubbed
Commit#closes_issues meaning that the only code that was really tested
was the call to Array#uniq to filter out duplicate issues. As this code
is no longer used (nor present) the corresponding tests were removed.

Related: gitlab-org/gitlab-ce#12419

This is the git default and will help to prevent false positive matches.

Closes #12706

This partially reverts 6a5cd3ca - we keep the migration and add a new
migration that reverts it in order to keep migration history intact.

By simply loading the first event from the already sorted set we save
ourselves extra (slow) queries just to get the latest update timestamp.
This removes the need for Event.latest_update_time and significantly
reduces the time needed to build an Atom feed.

Fixes gitlab-org/gitlab-ce#12415

The pattern in the `::reference_pattern` class method in the
ExternalIssue model does not match all valid forms of JIRA project
names. I have updated the regex to match JIRA project names with numbers
and underscores. More information on valid JIRA project names can be
found here:
https://confluence.atlassian.com/jira/changing-the-project-key-format-192534.html

* The first character must be a letter,
* All letters used in the project key must be from the Modern Roman Alphabet and upper case, and
* Only letters, numbers or the underscore character can be used.

By instead using a sub-query we save ourselves the overhead of loading
any data into memory only to pass it on to another query.

Leaving out the -I option in "git grep" would cause an Error 500 because the
resulting line would include "Binary file X matches"

Gitlab::Git::Repository#branch_count is a tad faster than the previous
setup. See gitlab-org/gitlab_git!62 for more information.

This also enable a project creator to add themselves as a master of the
project.