feat: support elide 1.6.0-rc.2 and make it the default

[Sam Gammon (sgammon)]

• Adds 1.6.0-rc.2 to versions.bzl (SRI from the published .sha256 assets) and makes it DEFAULT_VERSION. RCs and releases are tagged v<version>, so release_tag() maps a version to its tag for elide_download and the benchmark workflow; nightlies are unchanged.
• elide_download restores the exec bit on lib/jspawnhelper. The archive ships it 0644, so every subprocess started from elide java failed with posix_spawn … error: 13; this broke the KotlinBuilder shim's fallback to the stock builder.
• Fallback gated --sun-misc-unsafe-memory-access=allow on the shim's own JVM (Elide's, JDK 25) while the stock builder runs on the Bazel JDK (JAVA_HOME), which may be older: Unrecognized option on JDK 21. It now reads the feature version from $JAVA_HOME/release.

Verified locally on Bazel 9.2.0, linux-amd64, Elide 1.6.0-rc.2:

• bazel test //tests:all //tests/kotlin_builder:all: 37/37 pass
• e2e/integration: bazel build -- //... -//:native_app, //:HelloTest PASSED, //:app and //:kt_app run
• e2e/kotlin_builder: bazel build //:greeter //:annotated succeeds (failed before the two fixes above)
• tools/update_docs.sh regenerated docs/extensions.md

[Copilot]

🟡 Changes recommended

Release-tag translation currently alters custom URL template semantics and can break BYO artifacts or mirrors.

1 open finding

• Preserve raw version for custom URL templates · New

What changed in this PR

Adds Elide 1.6.0-rc.2 as the default runtime and addresses compatibility issues affecting subprocesses and Kotlin fallback builds.

Changes:

• Adds release-tag mapping and updates benchmark downloads.
• Restores jspawnhelper execution permissions.
• Detects the fallback builder’s JDK version from JAVA_HOME.

File	Description
elide/​private/​versions.bzl	Pins 1.6.0-rc.2 and maps its release tag.
elide/​private/​download.bzl	Uses release tags and fixes helper permissions.
elide/​kotlin/​builder/​Fallback.kt	Detects the fallback JDK feature version.
elide/​extensions.bzl	Updates version documentation.
tests/​versions_test.bzl	Tests tag mapping and default integrity.
tests/​kotlin_builder/​FallbackTest.kt	Tests JDK version parsing.
docs/​extensions.md	Regenerates extension documentation.
.github/​workflows/​benchmarks.yml	Downloads benchmarks from the mapped tag.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[codspeed]

Merging this PR will degrade performance by 3.11%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 2 improved benchmarks
❌ 2 regressed benchmarks
✅ 5 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	test_javac_full[elide]	22.1 ms	29.3 ms	-24.65%
❌	test_incremental[integration]	1.9 s	2.4 s	-19.42%
⚡	test_kotlinc_abi_only	198 ms	163.4 ms	+21.16%
⚡	test_kotlinc_full[elide]	191.1 ms	159.5 ms	+19.82%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing fix/elide-1.6 (b4744a4) with main (13539ef)}