Add markdown documention for the data profiling example

[sscdotopen]

Issue #9

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[stefan-grafberger]

Should we link the corresponding classes in the Readme version? We don't mention some of the advanced configuration options we offer here, at the same time this makes this example very short and easy to understand. However, should me maybe link the corresponding classes for the profiles (and mention the approximate percentiles) and for the RunBuilder?

Also, would it make sense to show the JSON version of a profile and mention the convenience function we offer for that conversion?

We could also consider adding examples on how to visualize this profiles, what do you think about this?

[sscdotopen]

It would be great to have a second example that shows the advanced options and gives hints on visualization. Could you create an issue for that?

Where exactly should we link add links? Each example always links to its scala file and we have a link to the example package in README.md already.

[stefan-grafberger]

Maybe something like this?

We have two different types of profiles we return, StandardColumnProfiles and NumericColumnProfiles. If the cardinality of a column is low enough, we also include a histogram with the StandardColumnProfile.

We also have more advanced configuration options as part of the ColumnProfilerRunner API, see here. We will also include an example of a more advanced use case in the future.

However, even without this, it should be quite easy to find these classes in the code.