fix(live-debugger): guard against NULL sidecar in diagnostics calls

[Leiyks]

Problem

The min install tests CI job was failing with a crash caused by a NULL dereference in ext/live_debugger.c. After commit 6b55c3ee5 introduced per-thread sidecar (DDTRACE_G(sidecar)), calls to ddog_send_debugger_diagnostics() pass &DDTRACE_G(sidecar) without checking for NULL. In minimal install environments, the sidecar hasn't been initialized by the time live debugger diagnostics are sent.

Fix

Add a NULL guard at each ddog_send_debugger_diagnostics() call site in ext/live_debugger.c. Use ddtrace_sidecar_for_signal (the global fallback sidecar handle) when the per-thread sidecar is NULL, and skip the call entirely if both are NULL.

Testing

The min install tests job runs with a stripped-down PHP environment where the sidecar isn't always initialized. This fix prevents the crash in that scenario.

[bwoebi]

This points to a deeper problem.
This testsuite never should have DDTRACE_G(sidecar) == NULL. How does that happen?

We never should install debugger probes before initializing the sidecar. And we probably should also skip installing them if connecting to the sidecar failed.

[datadog-datadog-prod-us1-2]

🎉 All green!

❄️ No new flaky tests detected
🧪 All tests passed

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 60.64% (-0.01%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 8de240b | Docs | Datadog PR Page | Give us feedback!}

[pr-commenter]

Benchmarks [ tracer ]

Benchmark execution time: 2026-04-24 12:13:44

Comparing candidate commit 8de240b in PR branch leiyks/fix-ci-live-debugger-null-sidecar with baseline commit b7eb412 in branch master.

Found 1 performance improvements and 2 performance regressions! Performance is the same for 191 metrics, 0 unstable metrics.

scenario:MessagePackSerializationBench/benchMessagePackSerialization-opcache

• 🟩 execution_time [-6.807µs; -4.913µs] or [-5.947%; -4.293%]

scenario:SamplingRuleMatchingBench/benchRegexMatching3

• 🟥 execution_time [+35.742ns; +91.858ns] or [+2.450%; +6.298%]

scenario:SamplingRuleMatchingBench/benchRegexMatching4

• 🟥 execution_time [+29.433ns; +96.967ns] or [+2.005%; +6.605%]