Update to make sql server bulk load verification non case-sensitive

Catch generic exceptions in main loop

update to force xml implementation

with default hadoop the com.sun.xml.* implementation is used which is compatible with spark-xml. with later hadoop versions the default is com.ctc.wstx.stax.* which will not work with spark-xml. this will force correct version

make release

make release

Update alpine base image to ensure APK exploit fixed

TensorFlowServing and UDF

- refactored `TensorFlowServingTransform` to call the `REST` API (in batches) which is easier due to library conflicts creating `protobuf`.
- create the `UDF` registration mechanism allowing logical extension point for common functions missing from Spark core.

Add a scala-sbt Dockerfile for predictable build

Update docs for DiffTransform to clarify persistence