Make negative Capybara matchers faster

Capybara matchers are smart enough to have these two statements
functionally equivalent:

    expect(page).not_to have_xpath('a')
    expect(page).to have_no_xpath('a')

But this only works if `Capybara::RSpecMatchers` module is included.
It is included automatically for specs in `spec/features`, but not for
pages objects.

Adding it speeds up tests relying on expecting something `not_to` be
present.

spec/support/pages/page.rb

@@ -29,6 +29,7 @@
 module Pages
   class Page
     include Capybara::DSL
+    include Capybara::RSpecMatchers
     include RSpec::Matchers
     include OpenProject::StaticRouting::UrlHelpers