feat: Android O11Y cost based buffering + JSON exporters

[abelonogov-ld]

Summary

• Reusing Cost Based Buffering for O11Y from SR (matching Swift)
• Drop Protobuf in favor JSON (matching Swift)

How did you test this change?

Are there any deployment considerations?

---

Note

Medium Risk
Replaces OpenTelemetry’s built-in OTLP exporters/processors with a new queued, cost-based batching pipeline and a custom OTLP/HTTP+JSON client, which could affect telemetry delivery, sampling, and flush semantics. Also changes the public Observe.flush() signature to non-blocking Unit, requiring downstream callers to update.

Overview
Switches Android observability export to a cost-based buffered queue shared across logs, traces, and metrics (via EventQueue + BatchWorker), aligning behavior with the Swift implementation.

Replaces OTLP Protobuf exporters with custom OTLP/HTTP+JSON exporters (OtlpHttpClient + per-signal JSON adapters/models) and rewires ObservabilityService to enqueue log/span/metric payloads (EventLogRecordProcessor, EventSpanProcessor, EventMetricExporter) while keeping debug/inspector exporters as optional delegates.

Updates Observe.flush() (and LDObserve) to be non-blocking and removes prior processor force-flush aggregation; adds Kotlin compiler/runtime pinning in Gradle and refactors gzip compression into GzipUtil, with new unit tests covering the new queueing/export/JSON encoding behavior.

^{Reviewed by Cursor Bugbot for commit 7f638be. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 095be57. Configure here.}