[Coral-Service] Add rewriteType field to translation endpoint

[KevinGe00]

What changes are proposed in this pull request, and why are they necessary?

Adding a rewrite field in the api/translations/translate endpoint to support new coral-service rewrite feature. This PR also modifies coral service to accept Spark as a source language.

How was this patch tested?

no rewrite param specified, returns a translation with no rewrite

curl --header "Content-Type: application/json" \
  --request POST \
  --data '{
    "fromLanguage":"hive", 
    "toLanguage":"trino", 
    "query":"SELECT * FROM db1.airport"
  }' \
  http://localhost:8080/api/translations/translate`

rewrite param is "none", returns a translation with no rewrite

curl --header "Content-Type: application/json" \
    --request POST \                                                                                                              
    --data '{                                  
    "sourceLanguage":"hive",
    "targetLanguage":"trino",
    "query":"SELECT * FROM db1.airport",
    "rewriteType":"none"
    }' \

valid rewrite param, returns translation with requested rewrite

curl --header "Content-Type: application/json" \
    --request POST \                                                                                                              
    --data '{                                  
    "sourceLanguage":"hive",
    "targetLanguage":"trino",
    "query":"SELECT * FROM db1.airport",
    "rewriteType":"incremental"
    }' \

invalid rewrite param, returns a translation with no rewrite

curl --header "Content-Type: application/json" \
    --request POST \                                                                                                              
    --data '{                                  
    "sourceLanguage":"hive",
    "targetLanguage":"trino",
    "query":"SELECT * FROM db1.airport",
    "rewriteType":"notavalidrewritetype"
    }' \

[wmoustafa]

no need for /input.

[wmoustafa]

How does this work when the source/target specified by the user is not Spark?

[KevinGe00]

getIncrementalQuery(String query, String sourceLanguage, String targetLanguage) takes in a sourceLanguage param.

[wmoustafa]

Still not quite clear. Why are not we using getIncrementalQuery here and leaving it as a TODO?

[KevinGe00]

In order to use getIncrementalQuery, we would need to /api/incremental/rewrite to take in a targetLanguage field in the request.

Since changes to this endpoint could break the existing DBT integration, it might be worth it to make this it's own separate effort especially since this is a more complicated endpoint to test.

[wmoustafa]

Can we hardcode "spark" and get rid of getSparkIncrementalQueryFromUserSql? The TODO will look different as well.

[wmoustafa]

Let us allow Spark and route it to Hive. Also you can say
Currently, only Hive, Spark, and Trino SQL are supported as source languages for Coral IR visualization

[KevinGe00]

Do we also do this for the translation endpoint?