* Basic memberships retrieval

* Change the way the memberships are saved in the client API database

* Retrieve single membership

* Get memberships only if the user is or has been in the room

* Check server name on room ID instead of user ID

* Save the join membership event and updates it when necessary

* Membership events retrieval + update on leave

* Implement the API on the roomserver and client API server

* Fix comments

* Remove the functions and attributes used before the new query API

* Explicitely state what we return in query

* Remove tab

* Add a HTTP API for roomserver input

* Fix integration tests

* Set depth of events and whether they need to be federated.

Set the depth of each new event to one greater than the maximum depth
of it's direct ancestors.

Add a flag to each event passing through the roomserver that tells us
whether the event needs to be sent over federation.

We do this by passing the name of the server to send the event as.
We will need this capability if we want to support vhosting as it is
not possible to tell from the event alone which server to send it as.

(The reason for this is that sometimes a event needs to be sent on
behalf of a different remote matrix server)

* Update roomserver integration tests

* More comments

* Constants and comments

* Fix comments

* Document the int64 returned by LatestEventIDs

This reverts commit ce82158a.

* Add API for querying events by ID.

* Fix tense

* Start implementing federation ingress

* More stuff

* Hook up federation event receiving

* Fix comments

* Comment on the order of the arrays

* Add API for querying the state after a list of events

* Implement state after events API

* Actually implement the API

* Reorder .travis.yml

* Comments

* Use common.MakeAPI

* Update gomatrixserverib dep

* Use helper functions when creating rooms

* Use gomatrixserverlib.StateKeyTuple

* Start implementing a query API for go using net/rpc

* Use a conventional JSON POST API rather than go net/rpc

net/rpc doesn't automatically handle reconnecting and we have better
logging and metrics infrastructure for monitoring HTTP apis.

* Implement the query API and add it to the integration tests

* Increase the timeout, travis seems to be a bit slow

* Clarify that state events are the things that are not returned if they are not requested

* Add utility function for converting arrays of numeric event IDs to pq Int64Arrays

* Warn people against requesting empty state keys by accident