feat: Add configurable flush interval

[marandaneto]

💡 Motivation and Context

PHP batching previously flushed queued events only when batch_size was reached, via explicit flush(), or during cleanup. This adds a configurable flush interval mechanism for low-volume queues while accounting for PHP's lack of a portable in-process background timer.

💚 How did you test it?

• php -l lib/QueueConsumer.php && php -l lib/Client.php && php -l lib/PostHog.php && php -l test/PostHogTest.php
• ./vendor/bin/phpunit test/PostHogTest.php --filter 'FlushInterval|testInvalidCaptureFlushIntervalDefaultsToFiveSeconds' --no-coverage
• php scripts/check-public-api.php
• git diff --check
• ! rg -n 'flush_interval_ms|"flush_interval"|flush_interval\?:' lib test README.md composer.json
• ./vendor/bin/phpunit --no-coverage — passes with existing warnings/deprecations

📝 Checklist

• I reviewed the submitted code.
• I added tests to verify the changes.
• I updated the docs if needed.
• No breaking change or entry added to the changelog.

If releasing new changes

• Ran pnpm changeset to generate a changeset file

🤖 Agent context

Autonomy: Human-driven (agent-assisted)

Implemented with Pi coding agent assistance. The change keeps the public API minimal with only flush_interval_seconds, defaults to 5 seconds, treats invalid, negative, or non-finite values as the default, flushes immediately when set to 0, and uses opportunistic interval flushing on the next enqueue because PHP does not provide a portable request-local background timer. Follow-up review feedback clarified the existing timeout and maximum_backoff_duration units in PHPDoc without renaming those options.

[greptile-apps]

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
test/PostHogTest.php:551-631
**Duplicate tests should be a single parameterised test**

`testNegativeCaptureFlushIntervalDefaultsToFiveSeconds` and `testInvalidCaptureFlushIntervalDefaultsToFiveSeconds` are byte-for-byte identical except for the value of `flush_interval_seconds` (`-1` vs `"1"`). The project rule says to prefer parameterised tests and to say everything OnceAndOnlyOnce. A single `@dataProvider` method covering both invalid inputs would satisfy both rules and make it trivial to add further edge cases (e.g. `NAN`, `INF`, `null`) without another copy-paste block.

_{Reviews (1): Last reviewed commit: "feat: add configurable flush interval" | Re-trigger Greptile}

[marandaneto]

#170 would be the ideal word but its much more effort

[greptile-apps]

_{Reviews (2): Last reviewed commit: "test: parameterize invalid flush interva..." | Re-trigger Greptile}