Fix emoji URLs in Markdown when relative_url_root is used

Path/URL helpers (e.g. `asset_url_helper`) already include the relative_url_root, so there is no need to attach it twice.

Also adds the ability to run rspecs with relative_url_defined on the enviornment. For example:

RELATIVE_URL_ROOT=/gitlab rspec

Closes #1728

See merge request !1254

Fix UI bugs with discussions
Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

See merge request !1265

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Do not let NGINX buffer Git HTTP requests

Before this change NGINX would convert a chunked HTTP POST (e.g.
git push) into a HTTP 1.0 single large POST. This creates an
unnecessary delay, and it creates unnecessary memory pressure on
gitlab-git-http-server.

For the response ('proxy_buffering') I am less sure that NGINX 's
buffering behavior is harmful, but it still makes more sense to me
not to interfere with gitlab-git-http-server (and the Golang net/http
server).

See merge request !1256

Add meta tag for referrer

https://gitlab.zendesk.com/agent/tickets/7054

Added a meta referrer tag, so that only the origin is sent to third party sites, instead of the entire URL, thus avoiding the leak of sensitive information like password reset tokens.

/cc @DouweM

See merge request !1263

Global labels

https://dev.gitlab.org/gitlab/gitlabhq/issues/2353

See merge request !1240

Fogbugz importer

Bump gitlab-shell to v2.6.5

[ci skip]

See merge request !1264

Added meta tag for referrer, so that only the origin is sent to third party sites, instead of the entire URL, thus avoiding the leak of sensitive information like password reset tokens.

Escape user-provided content in preserved HAML sections

See merge request !1927

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Issue redesign

* new issue UI
* fixed small font in issue/mr list

See merge request !1258

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Ui improvements

* fix project home page when activity set as default
* style commits, branches and tags pages to new style
* render flash message outside of content block
* style group pages

See merge request !1255

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Crowd integration

https://dev.gitlab.org/gitlab/gitlabhq/issues/2217

See merge request !1228

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Before this change NGINX would convert a chunked HTTP POST (e.g.
git push) into a HTTP 1.0 single large POST. This creates an
unnecessary delay, and it creates unnecessary memory pressure on
gitlab-git-http-server.

For the response ('proxy_buffering') I am less sure that NGINX 's
buffering behavior is harmful, but it still makes more sense to me
not to interfere with gitlab-git-http-server (and the Golang net/http
server).

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>

Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com>