[BUG] java.lang.UnsupportedOperationException: empty.minBy

Describe the bug
When running in EMR 7.10, I get this exception (run several times in different clusters):

25/12/24 08:52:15 INFO AppOptimizerAnalyzer: Generate Spark simulations with different sets of cores and max executors
25/12/24 08:52:17 INFO AppOptimizerAnalyzer: Generate application recommendations
StageID  Actual(ms)   Estimated(ms)  Tasks    AvgTask(ms)  MaxTask(ms)  Parents
-------------------------------------------------------------------------------
0        2225         1965           1        1965         1965
1        2258         2139           4        1403         2139
3        197          173            1        173          173          2
4        175          145            1        145          145
5        1358         1316           3        624          1316
6        481          422            4        347          422
8        177          147            1        147          147          7
9        4402         4333           4        1559         4333
10       469          440            4        423          440
12       57           32             1        32           32           11
13       130          110            1        110          110
14       510          486            3        210          486
Real Stage time      :   12s
Estimated Stage time :   11s
25/12/24 08:52:17 INFO AppOptimizerAnalyzer: Analyzing Optimizations for EC2 CostOpt
instance.emrPrice: 0.07348
runtimeHours: 0.015976666666666667
numInstances: 1
25/12/24 08:52:17 INFO AppOptimizerAnalyzer: Analyzing Optimizations for EC2 EfficiencyOpt
25/12/24 08:52:17 ERROR AppAnalyzer$: Failed in Analyzer AppOptimizerAnalyzer
java.lang.UnsupportedOperationException: empty.minBy
        at scala.collection.TraversableOnce.minBy(TraversableOnce.scala:305)
        at scala.collection.TraversableOnce.minBy$(TraversableOnce.scala:303)
        at scala.collection.AbstractTraversable.minBy(Traversable.scala:108)
        at com.amazonaws.emr.spark.optimizer.SparkEfficiencyOptimizer.recommend(SparkEfficiencyOptimizer.scala:57)
        at com.amazonaws.emr.spark.analyzer.AppOptimizerAnalyzer.$anonfun$analyze$8(AppOptimizerAnalyzer.scala:126)
        at scala.collection.immutable.List.foreach(List.scala:431)
        at com.amazonaws.emr.spark.analyzer.AppOptimizerAnalyzer.$anonfun$analyze$7(AppOptimizerAnalyzer.scala:117)
        at com.amazonaws.emr.spark.analyzer.AppOptimizerAnalyzer.$anonfun$analyze$7$adapted(AppOptimizerAnalyzer.scala:112)
        at scala.collection.immutable.List.foreach(List.scala:431)
        at com.amazonaws.emr.spark.analyzer.AppOptimizerAnalyzer.analyze(AppOptimizerAnalyzer.scala:112)
        at com.amazonaws.emr.spark.analyzer.AppAnalyzer.analyze(AppAnalyzer.scala:32)
        at com.amazonaws.emr.spark.analyzer.AppAnalyzer.analyze$(AppAnalyzer.scala:31)
        at com.amazonaws.emr.spark.analyzer.AppOptimizerAnalyzer.analyze(AppOptimizerAnalyzer.scala:56)
        at com.amazonaws.emr.spark.analyzer.AppAnalyzer$.$anonfun$start$1(AppAnalyzer.scala:66)
        at com.amazonaws.emr.spark.analyzer.AppAnalyzer$.$anonfun$start$1$adapted(AppAnalyzer.scala:64)
        at scala.collection.immutable.List.foreach(List.scala:431)
        at com.amazonaws.emr.spark.analyzer.AppAnalyzer$.start(AppAnalyzer.scala:64)
        at com.amazonaws.emr.spark.EmrSparkLogParser.analyze(EmrSparkLogParser.scala:33)
        at com.amazonaws.emr.SparkLogsAnalyzer$.delayedEndpoint$com$amazonaws$emr$SparkLogsAnalyzer$1(SparkLogsAnalyzer.scala:42)
        at com.amazonaws.emr.SparkLogsAnalyzer$delayedInit$body.apply(SparkLogsAnalyzer.scala:24)
        at scala.Function0.apply$mcV$sp(Function0.scala:39)
        at scala.Function0.apply$mcV$sp$(Function0.scala:39)
        at scala.runtime.AbstractFunction0.apply$mcV$sp(AbstractFunction0.scala:17)
        at scala.App.$anonfun$main$1$adapted(App.scala:80)
        at scala.collection.immutable.List.foreach(List.scala:431)
        at scala.App.main(App.scala:80)
        at scala.App.main$(App.scala:78)
        at com.amazonaws.emr.SparkLogsAnalyzer$.main(SparkLogsAnalyzer.scala:24)
        at com.amazonaws.emr.SparkLogsAnalyzer.main(SparkLogsAnalyzer.scala)
        at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
        at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
        at java.base/java.lang.reflect.Method.invoke(Method.java:569)
        at org.apache.spark.deploy.JavaMainApplication.start(SparkApplication.scala:52)
        at org.apache.spark.deploy.SparkSubmit.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:1150)
        at org.apache.spark.deploy.SparkSubmit.doRunMain$1(SparkSubmit.scala:200)
        at org.apache.spark.deploy.SparkSubmit.submit(SparkSubmit.scala:223)
        at org.apache.spark.deploy.SparkSubmit.doSubmit(SparkSubmit.scala:92)
        at org.apache.spark.deploy.SparkSubmit$$anon$2.doSubmit(SparkSubmit.scala:1246)
        at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:1255)
        at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)
25/12/24 08:52:17 INFO AppInsightsAnalyzer: Generate application insights...

To Reproduce
EMR release (e.g. emr-7.3.0) : 7.10.0
Steps to reproduce the behavior:

1. Create the cluster (runtime role, LF enabled, no Internet access, only VPC endpoints)
2. Run several steps
3. Login to master using SSM
4. Run the command:

spark-submit --class com.amazonaws.emr.SparkLogsAnalyzer   aws-emr-advisor-assembly-*.jar   --bucket xxxxx   hdfs:///var/log/spark/apps/application_xxxx_0003_1

5. The report is generated, although no cost/efficiency/performace data is show (all report numbers are 0)

Expected behavior
I would expect no exceptions and some data show in the cost/efficiency/performance sections.

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.

Start at SparkEfficiencyOptimizer.scala:57, where the stack trace reports empty.minBy, then trace how AppOptimizerAnalyzer.scala lines 112-126 reaches that recommendation. Reproduce with the supplied EMR 7.10.0 spark-submit command and event-log setup. Done means the analyzer completes without the exception and the cost, efficiency, and performance sections contain data.