Skip writing empty stages & skip loading stages without uid-s

[tovbinm]

Related issues
Follows up on #281.
Every stage must have a UID, if so some reason uid field is missing from the model json currently we fail miserably.

Describe the proposed solution
As precaution: 1) skip saving empty stages 2) skip reading stages without uid-s

[codecov]

Codecov Report

Merging #282 into master will increase coverage by 0.02%.
The diff coverage is 87.5%.

@@            Coverage Diff             @@
##           master     #282      +/-   ##
==========================================
+ Coverage   86.39%   86.42%   +0.02%     
==========================================
  Files         319      319              
  Lines       10448    10450       +2     
  Branches      353      555     +202     
==========================================
+ Hits         9027     9031       +4     
+ Misses       1421     1419       -2

Impacted Files	Coverage Δ
...cala/com/salesforce/op/OpWorkflowModelWriter.scala	100% <100%> (ø)	⬆️
...cala/com/salesforce/op/OpWorkflowModelReader.scala	94.44% <83.33%> (+0.1%)	⬆️
...es/src/main/scala/com/salesforce/op/OpParams.scala	89.79% <0%> (+4.08%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1b097d4...0ba2a49. Read the comment docs.

[wsuchy]

LGTM, as long as it is intended to silently omit stages without uid

[tovbinm]

@wsuchy every Spark org.apache.spark.ml.PipelineStage extends Params, which extends Identifiable which requires val uid: String - https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/ml/util/Identifiable.scala#L39