SchemaNotEqualError not showing the difference in metadata

[henrytomsf]

Related to #70 , when metadata is different between two schemas, the printout in the error logs don't explicitly show the difference in metadata since it only uses the __repr__ from pyspark's StructField.

[peterreckamp-8451]

Recently started using this library and this is the first issue I found, would love is this PR was reviewed and merged.

[MilkSilk]

+1 I encountered this problem today and took me half a day to figure out what was wrong. This issue is crucial due to 3 facts: chipsa not showing difference in metadata despite failing dataframe equality assertion, spark not containing metadata in str(df.schema) nor in df.printSchema() and lastly ignore_metadata parameter not being documented in the README.md. An easy improvement should be to add ignore_metadata into the README, which would probably help a lot of the developers that will encounter this issue in the future. In general chipsa should display the difference in metadata when the assertion is failing.