test(ci): exercise OpenResty LuaJIT in the Lua test matrix

[membphis]

Motivation

qjson's real deployment target is OpenResty (the bench harness uses resty; homepage is api7/lua-qjson). But the CI lua job installs Ubuntu's apt package luajit (upstream LuaJIT) and tests only against that.

lua-cjson's CI matrix explicitly includes a luajit-openresty dimension. OpenResty ships a patched LuaJIT fork whose GC behavior, ffi internals, and default flags can differ from upstream LuaJIT. Since qjson leans hard on FFI semantics and GC-pinning (_hold keeps the input buffer alive for the doc's lifetime — see CLAUDE.md "Buffer lifetime" invariant), divergence here is exactly the kind of thing a single-runtime CI would miss.

Proposed work

Add OpenResty's LuaJIT as a second dimension of the existing lua job (keep upstream LuaJIT too):

• Option A: leafo/gh-actions-lua with luaVersion: "luajit-openresty" (already used by lua-cjson).
• Option B: run the busted suite inside an openresty/openresty container, or install OpenResty and drive via resty.

Run the existing tests/lua/ busted suite (especially gc_spec.lua and the lazy-table / _hold GC-pinning paths) under both runtimes.

Acceptance

• CI lua job runs tests/lua/ under both upstream LuaJIT and OpenResty LuaJIT.
• GC-pinning / lifetime specs pass under OpenResty's LuaJIT.

Affected files

• .github/workflows/ci.yml

Distinct from #70 (OPM publishing / Windows) — this is about test coverage of the OpenResty runtime, not packaging/distribution.

[coderabbitai]

🔗 Related PRs

#43 - test: add cjson and simdjson fixture coverage [merged]
#45 - feat: release v0.1.0 [merged]
#54 - perf: Lua encode + cache + scanner + validation stack optimizations [merged]
#58 - docs: fix install instructions and qualify ARM64 performance claims [merged]
#78 - test(fuzz): add nesting-depth stack-safety target [closed]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!