Related to https://github.com/spantaleev/matrix-docker-ansible-deploy/pull/963

This also simplifies Prerequisites, which is great.

It'd be nice if we were doing these checks in some optional manner
and reporting them as helpful messages (using
`matrix_playbook_runtime_results`), but that's more complicated.
I'd rather drop these checks completely.

We don't really need to fail in such a spectactular way,
but it's probably good to do. It will only happen for people
who are defining their own user/group id, which is rare.

It seems like a good idea to tell them that this doesn't work
as they expect anymore and to ask them to remove these variables,
which otherwise give them a fake sense of hope.

Related to #486 (Github Pull Request).

If one runs the playbook with `--tags=setup-all`, it would have been
fine.

But running with a specific tag (e.g. `--tags=setup-riot-web`) would
have made that initialization be skipped, and the `matrix-riot-web` role
would fail, due to missing variables.

As suggested in #63 (Github issue), splitting the
playbook's logic into multiple roles will be beneficial for
maintainability.

This patch realizes this split. Still, some components
affect others, so the roles are not really independent of one
another. For example:
- disabling mxisd (`matrix_mxisd_enabled: false`), causes Synapse
and riot-web to reconfigure themselves with other (public)
Identity servers.

- enabling matrix-corporal (`matrix_corporal_enabled: true`) affects
how reverse-proxying (by `matrix-nginx-proxy`) is done, in order to
put matrix-corporal's gateway server in front of Synapse

We may be able to move away from such dependencies in the future,
at the expense of a more complicated manual configuration, but
it's probably not worth sacrificing the convenience we have now.

As part of this work, the way we do "start components" has been
redone now to use a loop, as suggested in #65 (Github issue).
This should make restarting faster and more reliable.