Use Java 26 for JavaDoc generation and remove custom dark-mode overrides

[liannacasper]

Motivation

• Java 26 ships native dark-mode support for javadoc, so rely on the toolchain instead of the repository's brittle faux dark-mode overrides.
• The website embeds generated JavaDocs, and the CI workflows that build/publish JavaDocs must use the same JDK to produce the new output.

Description

• Prefer JDK_26_HOME (fall back to JAVA_HOME) in ./.github/scripts/build_javadocs.sh so the JavaDoc tool comes from Java 26 when available.
• Update the JavaDoc-related GitHub Actions workflows (.github/workflows/javadocs.yml, .github/workflows/pr.yml, and .github/workflows/website-docs.yml) to install Java 26 and expose JDK_26_HOME where JavaDocs/site are built.
• Remove the repository's custom JavaDoc dark-mode variable overrides from docs/website/assets/css/extended/cn1-javadoc-embed.css so JavaDoc's native styling and dark mode can be used unmodified.
• The site build still embeds JavaDoc content via scripts/website/build.sh and the JavaDoc generation script; only the JDK/toolchain and CSS overrides were changed.

Testing

• Ran a shell syntax check on the modified scripts with bash -n .github/scripts/build_javadocs.sh scripts/website/build.sh, which completed successfully.
• Inspected the generated diffs to verify workflow JDK versions and CSS overrides were updated as intended (no runtime CI changes executed here).

---

Codex Task

[github-actions]

Cloudflare Preview

• URL: https://pr-4701-website-preview.codenameone.pages.dev
• Branch: pr-4701-website-preview

[github-actions]

✅ Continuous Quality Report

Test & Coverage

• ✅ Tests: 2363 total, 0 failed, 0 skipped
• 📊 Line coverage: 53.37% [HTML preview] [Download]
  • Lowest covered classes
    • com.codename1.ui.Display$EdtException – 0.00%
    • com.codename1.ui.plaf.CSSBorder$LinearGradient – 0.00%
    • com.codename1.util.EasyThread$InQueueRunnable – 0.00%
    • com.codename1.components.FloatingActionButton$ReleaseActionListener – 0.00%
    • com.codename1.io.Oauth2$ShowAuthenticationActionListener – 0.00%
    • com.codename1.io.Oauth2$RefreshTokenActionListener – 0.00%
    • com.codename1.util.EasyThread$RunAndWaitRunnable – 0.00%
    • com.codename1.components.ToastBar$FlushAnimationCallback – 0.00%
    • com.codename1.ui.plaf.CSSBorder$ColorStop – 0.00%
    • com.codename1.ui.ElevationComparator – 0.00%

Static Analysis

• SpotBugs [HTML preview] [Download]
  • ❌ ByteCodeTranslator: 109 findings (High: 2, Normal: 80, Low: 27)
    • Top findings
      • High: com/codename1/tools/translator/BytecodeMethod.java:559 – Method com.codename1.tools.translator.BytecodeMethod.findWritableFields(Set) seems to be useless (rule: UC_USELESS_VOID_METHOD)
      • High: com/codename1/tools/translator/bytecodes/TryCatch.java:91 – Write to static field com.codename1.tools.translator.bytecodes.TryCatch.hasTryCatch from instance method com.codename1.tools.translator.bytecodes.TryCatch.appendInstruction(StringBuilder, List) (rule: ST_WRITE_TO_STATIC_FROM_INSTANCE_METHOD)
      • Normal: com/codename1/tools/translator/ByteCodeClass.java:1802 – com.codename1.tools.translator.ByteCodeClass.getBaseClassObject() may expose internal representation by returning ByteCodeClass.baseClassObject (rule: EI_EXPOSE_REP)
      • …and 106 more
  • ❌ android: 413 findings (High: 24, Normal: 196, Low: 193)
    • Top findings
      • High: CodenameOne/src/com/codename1/impl/CodenameOneThread.java:48 – com.codename1.impl.CodenameOneThread.STACK_FRAME_SIZE isn't final but should be (rule: MS_SHOULD_BE_FINAL)
      • High: com/codename1/impl/android/AndroidImplementation.java:1085 – Write to static field com.codename1.impl.android.AndroidImplementation.context from instance method com.codename1.impl.android.AndroidImplementation.setContext(Context) (rule: ST_WRITE_TO_STATIC_FROM_INSTANCE_METHOD)
      • High: com/codename1/impl/android/AndroidImplementation.java:1286 – com.codename1.impl.android.AndroidImplementation.isDarkMode() has Boolean return type and returns explicit null (rule: NP_BOOLEAN_RETURN_NULL)
      • …and 410 more
  • ❌ codenameone-maven-plugin: 488 findings (High: 1, Normal: 315, Low: 172)
    • Top findings
      • High: com/codename1/maven/AbstractCN1Mojo.java:807 – com.codename1.maven.AbstractCN1Mojo.OS isn't final but should be (rule: MS_SHOULD_BE_FINAL)
      • Normal: com/codename1/builders/AndroidGradleBuilder.java:1018 – Dead store to googleServicesJson in com.codename1.builders.AndroidGradleBuilder.build(File, BuildRequest) (rule: DLS_DEAD_LOCAL_STORE)
      • Normal: com/codename1/builders/AndroidGradleBuilder.java:1043 – Possible null pointer dereference in com.codename1.builders.AndroidGradleBuilder.build(File, BuildRequest) due to return value of called method (rule: NP_NULL_ON_SOME_PATH_FROM_RETURN_VALUE)
      • …and 485 more
  • ❌ core-unittests: 1 findings (Normal: 1)
    • Top findings
      • Normal: CodenameOne/src/com/codename1/ui/util/MutableResource.java:239 – com.codename1.ui.util.MutableResource.saveL10N(DataOutputStream, Hashtable) makes inefficient use of keySet iterator instead of entrySet iterator (rule: WMI_WRONG_MAP_ITERATOR)
  • ❌ ios: 125 findings (High: 4, Normal: 75, Low: 46)
    • Top findings
      • High: com/codename1/impl/ios/IOSImplementation.java:1228 – com.codename1.impl.ios.IOSImplementation.isDarkMode() has Boolean return type and returns explicit null (rule: NP_BOOLEAN_RETURN_NULL)
      • High: com/codename1/impl/ios/IOSImplementation.java:128 – com.codename1.impl.ios.IOSImplementation.nativeInstance isn't final but should be (rule: MS_SHOULD_BE_FINAL)
      • High: com/codename1/impl/ios/IOSImplementation.java:214 – Write to static field com.codename1.impl.ios.IOSImplementation.instance from instance method com.codename1.impl.ios.IOSImplementation.init(Object) (rule: ST_WRITE_TO_STATIC_FROM_INSTANCE_METHOD)
      • …and 122 more
• ✅ PMD: 0 findings (no issues) [Report archive]
• ✅ Checkstyle: 0 findings (no issues) [Report archive]

Generated automatically by the PR CI workflow.