[SPARK-49153] Increase Gradle JVM memory to 4g like Spark repo

[dongjoon-hyun]

What changes were proposed in this pull request?

This PR aims to increase Gradle JVM memory to 4g like Spark repo.

• https://docs.gradle.org/current/userguide/performance.html#increase_the_heap_size

Why are the changes needed?

Unlike CI, in the local build, Gradle Daemon is reused. The compilation is unstable like the following because the default Gradle JVM memory is 512MB.

> Task :spark-operator:compileTestJava
Note: /Users/dongjoon/APACHE/spark-kubernetes-operator/spark-operator/src/test/java/org/apache/spark/k8s/operator/metrics/healthcheck/SentinelManagerTest.java uses or overrides a deprecated API.
Note: Recompile with -Xlint:deprecation for details.
Note: Some input files use unchecked or unsafe operations.
Note: Recompile with -Xlint:unchecked for details.
OpenJDK 64-Bit Server VM warning: CodeCache is full. Compiler has been disabled.
[1.372s][warning][codecache] CodeCache is full. Compiler has been disabled.
OpenJDK 64-Bit Server VM warning: Try increasing the code cache size using -XX:ReservedCodeCacheSize=
[1.372s][warning][codecache] Try increasing the code cache size using -XX:ReservedCodeCacheSize=
CodeCache: size=2944Kb used=2943Kb max_used=2943Kb free=0Kb
 bounds [0x0000000105004000, 0x00000001052e4000, 0x00000001052e4000]
 total_blobs=1102 nmethods=464 adapters=554
 compilation: disabled (not enough contiguous free space left)
              stopped_count=1, restarted_count=0
 full_count=1 

Does this PR introduce any user-facing change?

No. This is a dev-only change.

How was this patch tested?

Check the JVM setting via jps locally.

Was this patch authored or co-authored using generative AI tooling?

No.

[dongjoon-hyun]

cc @viirya and @jiangzho

[viirya]

Looks good to me.

[dongjoon-hyun]

Thank you, @viirya !
Merged to main.