Handle spark's `BinaryType`

[fishfood34]

Pull Request Description

Today BinaryType columns get written as a string by just using .toString on a byte array resulting in a value similar to [B@7fdcec67. This PR updates the handling of this column type to write a base64 encoded string instead.

---

Future Release Comment

[Add description of your change, to include in the next release] [Delete any or all irrelevant sections, e.g. if your change does not warrant a release comment at all]

Breaking Changes:

• None

Features:

• None

Fixes:

• BinaryType data ingestion

[fishfood34]

@microsoft-github-policy-service agree company="GitHub"

[ag-ramachandran]

Hello @fishfood34 , Thanks for the contribution. Looks good to me overall. Couple of points

• This will probably need an update on the docs that binary types by default will get converted to base64.

• You will need to bump the maven version for the release too (in the root POM)

[fishfood34]

@ag-ramachandran I believe I have made the requested changes. Please let me know if you see anything else I missed.