records: add 2016 pileup record

[katilp]

(closes #3569)

Adds the record for the 2016 pileup sample.

To-do:

• check the actual size from eos (it is not taken from DAS as this is a 10% fraction of the original 500 TB dataset)
• modify the template so that the related dataset field won't appear for PREMIX
  

The related dataset field might have come from some local cache, does not appear anymore.

[katilp]

@tiborsimko This was run with --ignore-eos-store, maybe we can look together that part and add the modifications to cernopendata/data-curation#207 accordingly.

[katilp]

commit 0805b20 gives the correct number of events and forces the category "Pileup"

[katilp]

Config files needed for the pileup data provenance are now in /eos/opendata/cms/upload/configs2016-pileup and can be moved to /eos/opendata/cms/configuration-files/MonteCarlo2016
This directory already contains config files of the data science samples released earlier.

[tiborsimko]

Thanks, added file information, fixed usage links and other minor metadata information such as publishing year.

(And FYI I've added the docker image build fix on top from PR #3602 so that CI would be green until PR #3597 fully works.)