Run Synapse workers in their own containers

This switches the `docker exec` method of spawning
Synapse workers inside the `matrix-synapse` container with
dedicated containers for each worker.

We also have dedicated systemd services for each worker,
so this are now:
- more consistent with everything else (we don't use systemd
instantiated services anywhere)
- we don't need the "parse systemd instance name into worker name +
port" part
- we don't need to keep track of PIDs manually
- we don't need jq (less depenendencies)
- workers dying would be restarted by systemd correctly, like any other
service
- `docker ps` shows each worker separately and we can observe resource
usage

roles/matrix-synapse/files/usr-local-bin/matrix-synapse-worker-write-pid

@@ -1,30 +0,0 @@
-#!/bin/bash
-# Find a synapse worker's PID and write it to a file so systemd can manage it as a service
-
-# example invocation:
-# matrix-synapse-worker-write-pid user_dir:18700 /run/matrix-synapse-worker.user_dir:18700.pid
-
-docker_api_call() { curl --silent --unix-socket /var/run/docker.sock ${@}; }
-
-TARGETCONTAINER=matrix-synapse
-TARGETWORKER=${1}
-PIDFILE=${2}
-
-# get ID list of subprocesses executed in $TARGETCONTAINER, and for each..
-for EXECID in $(docker_api_call http://localhost/containers/${TARGETCONTAINER}/json | jq --raw-output '.ExecIDs[]')
-do
-  # fetch detailed process info
-  EXECINFO=$(docker_api_call http://localhost/exec/${EXECID}/json)
-
-  # extract config file path from last command argument
-  WORKERCONFIGFILE=$(echo ${EXECINFO} | jq --raw-output .ProcessConfig.arguments[-1])
-
-  # reconstruct worker name
-  WORKERNAME=${WORKERCONFIGFILE#*/worker.}
-  WORKERNAME=${WORKERNAME%.yaml}
-
-  # if name matches the target worker: write out most recent PID & quit
-  [ "${WORKERNAME}" = "${TARGETWORKER}" ] \
-    && echo ${EXECINFO} | jq --raw-output .Pid > ${PIDFILE} \
-    && exit 0
-done