Drop Spark BigTable version to unlock DataProc submission (#167)

## Summary
While testing the Flink side of things, I noticed that our DataProc
submission was broken due to some opencensus version mismatches:
```
[info]   com.google.common.util.concurrent.ExecutionError: java.lang.NoSuchMethodError: 'io.opencensus.tags.TagContext io.opencensus.tags.unsafe.ContextUtils.getValue(io.grpc.Context)'
[info]   at com.google.common.util.concurrent.Futures.wrapAndThrowUnchecked(Futures.java:1387)
[info]   at com.google.common.util.concurrent.Futures.getUnchecked(Futures.java:1380)
[info]   at com.google.api.gax.rpc.ApiExceptions.callAndTranslateApiException(ApiExceptions.java:53)
[info]   at com.google.api.gax.rpc.UnaryCallable.call(UnaryCallable.java:112)
[info]   at com.google.cloud.dataproc.v1.JobControllerClient.submitJob(JobControllerClient.java:435)
[info]   at com.google.cloud.dataproc.v1.JobControllerClient.submitJob(JobControllerClient.java:404)
[info]   at ai.chronon.integrations.cloud_gcp.DataprocSubmitter.submit(DataprocSubmitter.scala:70)
[info]   at ai.chronon.integrations.cloud_gcp.test.DataprocSubmitterTest.$anonfun$new$4(DataprocSubmitterTest.scala:77)
[info]   at scala.runtime.java8.JFunction0$mcV$sp.apply(JFunction0$mcV$sp.java:23)
[info]   at org.scalatest.OutcomeOf.outcomeOf(OutcomeOf.scala:85)
[info]   ...
[info]   Cause: java.lang.NoSuchMethodError: 'io.opencensus.tags.TagContext io.opencensus.tags.unsafe.ContextUtils.getValue(io.grpc.Context)'
[info]   at io.opencensus.implcore.tags.CurrentTagMapUtils.getCurrentTagMap(CurrentTagMapUtils.java:37)
[info]   at io.opencensus.implcore.tags.TaggerImpl.getCurrentTagContext(TaggerImpl.java:51)
[info]   at io.opencensus.implcore.tags.TaggerImpl.getCurrentTagContext(TaggerImpl.java:31)
[info]   at io.grpc.census.CensusStatsModule$StatsClientInterceptor.interceptCall(CensusStatsModule.java:801)
[info]   at io.grpc.ClientInterceptors$InterceptorChannel.newCall(ClientInterceptors.java:156)
[info]   at com.google.api.gax.grpc.GrpcChannelUUIDInterceptor.interceptCall(GrpcChannelUUIDInterceptor.java:52)
[info]   at io.grpc.ClientInterceptors$InterceptorChannel.newCall(ClientInterceptors.java:156)
[info]   at com.google.api.gax.grpc.GrpcHeaderInterceptor.interceptCall(GrpcHeaderInterceptor.java:80)
[info]   at io.grpc.ClientInterceptors$InterceptorChannel.newCall(ClientInterceptors.java:156)
[info]   at com.google.api.gax.grpc.GrpcMetadataHandlerInterceptor.interceptCall(GrpcMetadataHandlerInterceptor.java:54)
[info]   ...
```

Tried playing with different opencensus versions + lib management etc
and @tchow-zlai pointed out that this was due to the spark-bigtable dep.
We don't need this dep for the mainstream as we're using the BQ export
data query to load data into BT. The Spark-Bt connector is required
primarily for docker quickstart testing. So in this PR, I've yanked the
dep to re-enable the dataproc submission and reworked the load-data
script and Dockerfile to download the spark-bigtable + slf4j jars to
pass them in only for the Spark2bigTableLoader app in the docker setup.

## Checklist
- [ ] Added Unit Tests
- [ ] Covered by existing CI
- [X] Integration tested
- [ ] Documentation update



<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

- **New Features**
- Updated Docker image to include custom JAR files for BigTable and
logging
	- Modified Spark configuration to use updated Bigtable dependencies
	- Enhanced data loading script with improved JAR file handling

- **Dependency Updates**
- Replaced Spark-specific BigTable connector with general Bigtable HBase
library
	- Updated Google Cloud library dependencies to latest versions

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

Author: piyush-zlai

build.sbt

@@ -217,7 +217,6 @@ lazy val cloud_gcp = project
     libraryDependencies += "com.google.cloud.bigdataoss" % "gcsio" % "3.0.3", // need it for https://github.com/GoogleCloudDataproc/hadoop-connectors/blob/master/gcsio/src/main/java/com/google/cloud/hadoop/gcsio/GoogleCloudStorageFileSystem.java
     libraryDependencies += "io.circe" %% "circe-yaml" % "1.15.0",
     libraryDependencies += "com.google.cloud.spark" %% s"spark-bigquery-with-dependencies" % "0.41.0",
-    libraryDependencies += "com.google.cloud.spark.bigtable" %% "spark-bigtable" % "0.2.1",
     libraryDependencies += "com.google.cloud.bigtable" % "bigtable-hbase-2.x" % "2.14.2",
     libraryDependencies ++= circe,
     libraryDependencies ++= avro,

quickstart/cloud_gcp/scripts/load_data.sh

@@ -33,7 +33,9 @@ echo "GroupBy upload batch jobs completed successfully!"
 
 echo "Uploading tables to KV Store"
 for dataset in purchases returns; do
-  if ! spark-submit --driver-class-path "$CLASSPATH" --class ai.chronon.integrations.cloud_gcp.Spark2BigTableLoader \
+  if ! spark-submit --driver-class-path "$CLASSPATH:/opt/custom-jars/*" \
+    --jars "/opt/custom-jars/spark-bigtable_2.12-0.2.1.jar,/opt/custom-jars/log4j-slf4j-impl-2.20.0.jar" \
+    --class ai.chronon.integrations.cloud_gcp.Spark2BigTableLoader \
     --master local[*] $CLOUD_GCP_JAR --table-name default.quickstart_${dataset}_v1_upload --dataset quickstart.${dataset}.v1 \
     --end-ds 2023-11-30 --project-id $GCP_PROJECT_ID --instance-id $GCP_INSTANCE_ID; then
     echo "Error: Failed to upload table to KV Store" >&2