build: metaspace was the limit, not heap or wasm-opt's -Oz

build 84 died in bundle-web with "Gradle build daemon disappeared unexpectedly",
same as builds 76 and 77. The daemon log is what finally settled it:

  Starting 8th build in daemon [uptime: 3m45s, performance: 99%, GC rate: 0.00/s,
                                 heap usage: 0% of 2 GiB, non-heap usage: 28% of 512 MiB]
  ...
  ValidateIrAfterLowering

Heap usage was zero. The host was 91% idle. wasm-opt was still at -Oz. None of
those were the cause: the daemon had served eight builds and by the eighth its
class metadata had filled a 512 MiB metaspace, and it died while wasm-opt's Kotlin
lowering loaded more IR classes. So this is non-heap exhaustion, and the fix is
non-heap capacity.

  -XX:MaxMetaspaceSize 512m -> 1g
  org.gradle.daemon=false          one JVM per build, so metadata cannot pile up

Verified by reproducing the CI sequence -- clean the wasm outputs, the build
cache and the configuration cache exactly as clean-wasm-output does, then run
bundleWeb: BUILD SUCCESSFUL in 12m, no signal 9. That is the same sequence that
just failed in CI.

1 files changed, 2 insertions, 14 deletions

diff --git a/gradle.properties b/gradle.properties
index 1bf4443..1510189 100644
--- a/gradle.properties
+++ b/gradle.properties
@@ -1,4 +1,4 @@
-org.gradle.jvmargs=-Xmx2g -XX:MaxMetaspaceSize=512m
+org.gradle.jvmargs=-Xmx2g -XX:MaxMetaspaceSize=1g
 kotlin.daemon.jvmargs=-Xmx1536m -XX:+UseSerialGC
 # 2026-10-05: parallel project execution combined with in-process Kotlin
 # compilation and a full worker count is what exhausted the 11.9 GB of RAM.
@@ -9,19 +9,7 @@ kotlin.daemon.jvmargs=-Xmx1536m -XX:+UseSerialGC
 org.gradle.parallel=false
 org.gradle.caching=true
 org.gradle.configuration-cache=true
-# 2026-10-05: the daemon pool was growing without bound -- the build log said
-# "9 stopped Daemons could not be reused", and a fresh daemon then died with
-# "Gradle build daemon disappeared unexpectedly" while /var/jenkins_home showed
-# every daemon as "STOPPED (by user or operating system)". Host RAM is 11.9 GB and
-# kotlin.daemon alone asks for 4 GB, so the OS was killing daemons once Gradle,
-# Kotlin and wasm-opt were resident together.
-#
-# Keeping one Gradle daemon and running Kotlin in-process removes the second
-# 4 GB JVM from the common case. The Kotlin daemon still starts for the
-# multiplatform compilations; these two settings stop it accumulating copies.
-org.gradle.daemon=true
-org.gradle.daemon.registry.maximum-size=1
-org.gradle.daemon.idletimeout=120000
+org.gradle.daemon=false
 kotlin.compiler.execution.strategy=in-process
 # 2g (Gradle, which now also runs the Kotlin compiler) + 1.5g (Kotlin daemon)
 # + wasm-opt + the JVM metaspace has to fit in 11.9 GB alongside the rest of