Name already in use

…oint_container into laboratory

Now the container correctly works after docker-compose down
and docker-compose up cycle.

1) fixed a couple of scripts by adding #! declaration,
2) temporarily disabled "targets" container - it was not used anyway,
3) added "resolveip fix" to grouper_data Dockerfile (as suggested
on the Slack channel),
4) added 2 new archetypes + added visualization of existing ones
in demo/complex2s,
5) removed "add focus" synchronization in LDAP resource,
6) other minor improvements.

These are done by Ivan Noris.

…nt_container into laboratory

1. Membership and group deletion is now handled correctly
2. Operations on large LDAP groups should now be faster
(but still not optimal)

Patch 1

Switch to Java 11

Here we fixed syntax errors + minor variable naming issues.

1) Removed multiple intents from LDAP resource, making operations faster
2) Replaced template + metarole combination (for each group type)
with single archetype.
3) Factored out common code to 2 archetype metaroles.

It failed with bucketed imports.

It is 389ds not OpenLDAP :)

1) Implemented filtering in ScriptedSQL search operation
2) Turned off recompilation of Groovy scripts on each DB request

These packages are part of 'directory' service in various scenarios.
It looks like there are some issues with current versions of these
packages. So I have temporarily removed them from demo/complex2s
scenario to make it buildable.

The processing path of Grouper groups was significantly changed:
Instead of pushing member information through ShadowType to OrgType
we keep it only in ShadowType (in index-only form).

Auto-updating users after membership change is currently disabled.

This is the same as demo/complex2 but using simplified
access to the Grouper resource (fetching groups with member lists
instead of members + groups).

Fix shibd startup