NoSuchMethodError: descriptors$EnumValueDescriptor when deploying Beam Java SDK 2.10.0 to Dataflow

My working environment:
* Apache Beam Java SDK version: works with 2.9.0 but failed with 2.10.0
* Runner: failed with both Direct Runner and Dataflow Runner
* Application code: Scala (note I did not use Scio)

I tried to change Apache Beam Java SDK version from 2.9.0 to 2.10.0 and deploy it to Dataflow but I got this error. It works with 2.9.0. Am I missing something?
```

Exception in thread "main" java.lang.NoSuchMethodError: org.apache.beam.model.pipeline.v1.RunnerApi$BeamConstants$Constants.getValueDescriptor()Lorg/apache/beam/vendor/grpc/v1p13p1/com/google/protobuf/Descriptors$EnumValueDescriptor;
at
org.apache.beam.sdk.transforms.windowing.BoundedWindow.extractTimestampFromProto(BoundedWindow.java:84)
at
org.apache.beam.sdk.transforms.windowing.BoundedWindow.(BoundedWindow.java:49)
at org.apache.beam.sdk.coders.CoderRegistry$CommonTypes.(CoderRegistry.java:140)
at
org.apache.beam.sdk.coders.CoderRegistry$CommonTypes.(CoderRegistry.java:97)
at org.apache.beam.sdk.coders.CoderRegistry.(CoderRegistry.java:160)
at
org.apache.beam.sdk.Pipeline.getCoderRegistry(Pipeline.java:326)
at org.apache.beam.sdk.io.kafka.KafkaIO$Read.expand(KafkaIO.java:707)
at
org.apache.beam.sdk.io.kafka.KafkaIO$Read.expand(KafkaIO.java:309)
at org.apache.beam.sdk.Pipeline.applyInternal(Pipeline.java:537)
at
org.apache.beam.sdk.Pipeline.applyTransform(Pipeline.java:488)
at org.apache.beam.sdk.values.PBegin.apply(PBegin.java:56)
at
org.apache.beam.sdk.Pipeline.apply(Pipeline.java:182)

```

My code is in Scala but it works well with Beam 2.9.0. 
```

val p = Pipeline.create(options)
p.apply(s"${bu.name}_ReadFromKafka", KafkaIO.read()
.withBootstrapServers(options.getBootstreapServers)

.updateConsumerProperties(config)
.withTopics(util.Arrays.asList(topicName))
.withKeyDeserializer(classOf[LongDeserializer])

.withValueDeserializer(classOf[StringDeserializer])
.withConsumerFactoryFn(
new KafkaTLSConsumerFactory(

projectId, options.getSourceBucket, options.getTrustStoreGCSKey, options.getKeyStoreGCSKey)))

.apply(s"${bu.name}_Convert", ParDo.of(new ConvertJSONTextToEPCTransaction(bu)))
.apply(s"${bu.name}_WriteToBQ",
BigQueryIO.write()
.to(bqDestTable)
.withSchema(schema)
.withFormatFunction(new
ConvertMessageToTable())
.withCreateDisposition(BigQueryIO.Write.CreateDisposition.CREATE_NEVER)

.withWriteDisposition(BigQueryIO.Write.WriteDisposition.WRITE_APPEND))
}
p.run

```

According to the error log, it failed at this part.
[https://github.com/apache/beam/blob/v2.10.0/sdks/java/core/src/main/java/org/apache/beam/sdk/transforms/windowing/BoundedWindow.java#L81-L85](https://github.com/apache/beam/blob/v2.10.0/sdks/java/core/src/main/java/org/apache/beam/sdk/transforms/windowing/BoundedWindow.java#L81-L85)
```

private static Instant extractTimestampFromProto(RunnerApi.BeamConstants.Constants constant) {

return new Instant(
Long.parseLong(
constant.getValueDescriptor().getOptions().getExtension(RunnerApi.beamConstant)));

}

```

This constant come from this part.
[https://github.com/apache/beam/blob/v2.10.0/sdks/java/core/src/main/java/org/apache/beam/sdk/transforms/windowing/BoundedWindow.java#L48-L49](https://github.com/apache/beam/blob/v2.10.0/sdks/java/core/src/main/java/org/apache/beam/sdk/transforms/windowing/BoundedWindow.java#L48-L49)
```

public static final Instant TIMESTAMP_MIN_VALUE =
extractTimestampFromProto(RunnerApi.BeamConstants.Constants.MIN_TIMESTAMP_MILLIS);

```

Imported from Jira [BEAM-6884](https://issues.apache.org/jira/browse/BEAM-6884). Original Jira may contain additional context.
Reported by: yohei.