Fix: Corrects $SPARK_HOME errors in spark container

[brockneedscoffee]

When installing spark via miniconda the spark home is not set so you cannot run spark in AML. You will get errors when you try to set the spark context or if you set the framework to pyspark in an AML pipeline step. This PR installs spark, sets the home, and installs open mpi dependency.

[welcome]

💖 Thanks for opening your first pull request! 💖 We use semantic commit messages to streamline the release process. Before your pull request can be merged, you should make sure your first commit and PR title start with a semantic prefix. This helps us to create release messages and credit you for your hard work!
Examples of commit messages with semantic prefixes:

• fix: Fix LightGBM crashes with empty partitions
• feat: Make HTTP on Spark back-offs configurable
• docs: Update Spark Serving usage
• build: Add codecov support
• perf: improve LightGBM memory usage
• refactor: make python code generation rely on classes
• style: Remove nulls from CNTKModel
• test: Add test coverage for CNTKModel

Make sure to check out the developer guide for guidance on testing your change.

[mhamilton723]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[codecov]

Codecov Report

Merging #1009 (a17834c) into master (2c223f6) will increase coverage by 0.35%.
The diff coverage is n/a.

❗ Current head a17834c differs from pull request most recent head 0040785. Consider uploading reports for the commit 0040785 to get more accurate results

@@            Coverage Diff             @@
##           master    #1009      +/-   ##
==========================================
+ Coverage   84.13%   84.49%   +0.35%     
==========================================
  Files         201      199       -2     
  Lines        9306     9172     -134     
  Branches      554      543      -11     
==========================================
- Hits         7830     7750      -80     
+ Misses       1476     1422      -54     

Impacted Files	Coverage Δ
...om/microsoft/ml/spark/train/AutoTrainedModel.scala	50.00% <0.00%> (-35.72%)	⬇️
.../org/apache/spark/ml/param/UntypedArrayParam.scala	37.50% <0.00%> (-20.40%)	⬇️
...n/scala/com/microsoft/ml/spark/stages/Lambda.scala	80.00% <0.00%> (-13.34%)	⬇️
...a/com/microsoft/ml/spark/io/http/HTTPClients.scala	76.66% <0.00%> (-6.67%)	⬇️
...icrosoft/ml/spark/featurize/CleanMissingData.scala	88.88% <0.00%> (-4.96%)	⬇️
.../com/microsoft/ml/spark/stages/ClassBalancer.scala	81.81% <0.00%> (-4.85%)	⬇️
...icrosoft/ml/spark/automl/TuneHyperparameters.scala	73.91% <0.00%> (-4.35%)	⬇️
.../microsoft/ml/spark/vw/VowpalWabbitRegressor.scala	70.00% <0.00%> (-2.73%)	⬇️
.../com/microsoft/ml/spark/automl/FindBestModel.scala	85.00% <0.00%> (-1.54%)	⬇️
.../microsoft/ml/spark/core/utils/ModelEquality.scala	85.71% <0.00%> (-1.25%)	⬇️
... and 71 more

---

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 2c223f6...0040785. Read the comment docs.

[brockneedscoffee]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[brockneedscoffee]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[brockneedscoffee]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mhamilton723]

@brockneedscoffee - looks good but we dont test docker notebook as part of the build yet so could you manually test these out?

Also could you group all the run steps together into a single run step and put ENV stuff in top of docker. Putting all of the run steps together (and deleting ant .tgzs you download in these steps) makes the docker image considerably smaller thanks!

[mhamilton723]

Group run steps together for smaller docker image