* refactor GetDirectChannel and CreateDirectChannel in one function
* remove CreateDirectChannel plugin api and update GetDirectChannel and GetGroupChannel plugin api
* update tests
We have constants for (most) of the notify props keys. We should use
them consistently, instead of sometimes and other times having strings
scattered around the place.
If some, but not all, notify props are specified for a user in the bulk
import data, and that is a newly created user, we must explicitly
initialise all the other notify props to their default values to avoid
breaking client assumptions.
* api4: improved error handling
* system_store: more logs
* integrate go-junit-report into test-te/test-ee
* add CI_MINIO_HOST and CI_INBUCKET_HOST instead of CI_HOST
* comment re: minio configuration issue
* fix TestStartServerPortUnavailable to pass even when root can bind to :21
* skip TestFindManifest_FolderPermission while running as root
* Stop inserting hashed plugin keys
* Address comments
* Remove else statement
* Address comments
* Add comment
* Update as per comments
* Refactor as per comments
* Update plugin_key_value_store.go
minor comment tweaks
* fail on non StatusNotFound for original key query
* idiomatic error handling, and do not clean up if insert fails
* [MM - 12370] Add cli command "webhook delete"
* Replace spaces with tabs
* Call cli command to delete hooks in tests
* Capture webhook object from create webhook to get webhook id
* Print error message when webhook not deleted
* Remove redundant error message and if condition in webhook delete test
* Fix build
* MM-11434 Only call PreparePostForClient once when creating a post
* Have PreparePostForClient provide new metadata when a post already has it and update tests
* fix webconn close semantics
Avoid race conditions in WebConn on shutdown by closing channels to guarantee all readers are notified. Wrap this with sync.Once to avoid closing the channel more than once.
* web_hub_test.go
* webhub: fix race condition on shutdown
Ensure that if the webhub shuts down in the process of sending, the caller unblocks given that the webhub will no longer consume incoming events.
* panic if app shutdown takes >30 seconds
* simplify WebConn::Pump channel semantics too
