* Fixed `url()` returning wrong path if using stream with grav root path in it, eg: `user-data://shop` when Grav is in `/shop`
* Fixed `url()` not returning a path to non-existing file (`user-data://shop` => `/user/data/shop`) if it is set to fail gracefully
* Fixed `url()` returning false on unknown streams, such as `ftp://domain.com`, they should be treated as external URL
* Fix bug, method init must be protected
* Added phpstan level 0
* Added exclusions
* Fix: incorrect case in Twig_SimpleFunction
* Fix, no abstract properties, if properties used in methods let's set them
* Fixed Psr\SimpleCache\InvalidArgumentException referenced with incorrect case
* added exclusions
* Fixed: Access to an undefined property Grav\Framework\Form\FormFlash::$uploadObjects.
* Fixed: does not call parent constructor from Grav\Common\Iterator.
* Fixed: does not call parent constructor from Grav\Common\Iterator.
* Fixed: does not call parent constructor
* Fixed: does not call parent constructor
* Minor: correct type for inflector
* Moved phpstan.neon out of public directory
* Added exclusion
* set GRAV_USER_INSTANCE to prevent LogiException in User
* Exlude Stream:create
* Minor: Missing storage property
* Minor: Fixed missing properties (phpstan level 1)
* Added type for $data
* Minor: Ensure $langs is initialized
* Fix possible bug in $http_response_header status code retrieval (PHP 7.1 only)
* Added exclusion for $http_response_header (isset is actually not required)
* Strict null check
* initial commit of refactored class structure
* more refactoring
* more progress
* more refactoring…
* Rendering stuff!
* Updates
* filter + sort
* inline stuff now added
* continued compatibility improvements + some test fixes
* more compatibility fixes
* CSS pipeline working!
* pipeline in the house
* Various fixes for testing bugs
* Fix timestamps
* CSS/JS Inline support
* Tidy
* Optimizations…
* fix for tests
* Added a couple of inline tests
* Move fix for #1114 to Truncator::truncateLetters
The original fix provided by #1125 in Utils::truncateHtml compared the
summary size with the full HTML string length.
Thus, the string was still being truncated (and the HTML rewritten) even
when only the HTML string length, and not the text length, exceeded the
summary size.
* Add fix for #1114 also to Truncator::truncateWords
