Work around preloaded enums being broken on early PHP 8.4 versions with observers

[bwoebi]

See also php/php-src#17715.

Relatively simple fix: just initialize the cache slots on every request. Has a slight per request-overhead, but avoids crashes when SomePreloadedEnum::cases().

Not adding a test, it's covered by preload_enum_observed.phpt from php-src.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 73.04%. Comparing base (368a541) to head (0590f35).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff            @@
##             master    #3093   +/-   ##
=========================================
  Coverage     73.04%   73.04%           
  Complexity     2792     2792           
=========================================
  Files           139      139           
  Lines         15284    15284           
  Branches       1047     1047           
=========================================
  Hits          11164    11164           
  Misses         3567     3567           
  Partials        553      553           

Flag	Coverage Δ
appsec-extension	68.28% <ø> (ø)
tracer-php	74.86% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 368a541...0590f35. Read the comment docs.

[pr-commenter]

Benchmarks [ tracer ]

Benchmark execution time: 2025-02-20 14:29:31

Comparing candidate commit 0590f35 in PR branch bob/init-enum-slots with baseline commit 368a541 in branch master.

Found 1 performance improvements and 1 performance regressions! Performance is the same for 176 metrics, 0 unstable metrics.

scenario:PDOBench/benchPDOBaseline

• 🟥 execution_time [+14.701µs; +18.056µs] or [+8.025%; +9.856%]

scenario:TraceFlushBench/benchFlushTrace

• 🟩 execution_time [-1000.000ns; -1000.000ns] or [-50.000%; -50.000%]