We'd like to test the hypothesis that having the updater terminate after only 100 loops is necessary. In the past we believed that it may be due to having OOM issues after a certain time.

Currently we have the issue that, since it's deployed as a deployment, terminating after a certain number of loops is considered a "crash". The then results in it entering a crashloop back off after a certain number of loops.

Instead let us deploy this to run "indefinitely".

See: https://github.com/wbstack/queryservice-updater/blob/ed284dc2b8401b5c15269d90637e9142384c38c6/src/main/java/org/wikidata/query/rdf/tool/WbStackUpdate.java#L138 for where looping occurs.

We should probably still allow setting the loop limit but remove it as required. When usnset (or negative perhaps?). It should run indefinitely