reuse encoders for ordinal variables [type:bug]

[bvanbreugel]

Description

Fixes #250 (comment)

Notes

• Only quick fix. Sklearn recommends against using LabelEncoder for features, I have not changed this.
• Results in error when there are unknown feature classes in e.g. X_syn (compared to X_gt). This is safer, though does mean we cannot evaluate easily on a very small real dataset. If we want to avoid this, we could use OrdinalEncoder instead, which allows ignore_unknown argument to be passed

Checklist

• I have followed the Contribution Guidelines and Code of Conduct
• I have commented my code following the van der Schaar Lab Styleguide
• I have labelled this PR with the relevant Type labels
• My changes are covered by tests

[robsdavis]

Thanks for having a look at this issue, Boris. But as you said you don't have time to get these tests passing. I am closing this PR and not merging. I will leave the related issue open and will review it when I get a chance.