distributed RDataFrame cannot process models passed via RReader

[vepadulano]

Check duplicate issues.

• Checked for duplicates

Description

See this user forum post https://root-forum.cern.ch/t/evaluating-mva-within-distr-rdf/63079

The issue is that the RReader class is not serializable, so we cannot send instances of the class to the workers when using distributed RDataFrame.

Reproducer

No minimal reproducer yet, but the forum post is quite clear.

ROOT version

Any

Installation method

Any

Operating system

Any

Additional context

No response

[dpiparo]

Making the class serialisable should not be hard, given its layout - come care might be required since the class is still in the Experimental namespace (might not be a big problem, but it depends on the migration strategy).
@lmoneta