feat(sdk): add otel.sdk.processor.log.processed self-diagnostics metric

[cijothomas]

Adds the otel.sdk.processor.log.processed counter to BatchLogProcessor, following the OTel SDK metrics semantic conventions.

Gated behind the existing experimental_metrics_bound_instruments feature flag — bound instruments make this practical on the hot path (~1.8 ns per emit vs ~50 ns with unbound).

Changes:

• BatchLogProcessor emits a counter on every emit() call with three attribute variants: success, queue_full, and processor_shutdown
• OTLP examples reordered to set up MeterProvider first (so the processor gets a real meter)
• Shutdown order corrected to tracer → logger → meter
• Design doc added at docs/design/observability.md
• Ignored test added (runs in isolation in CI via test.sh)

Ordering requirement: global::set_meter_provider() must be called before creating the LoggerProvider. Rust's global::meter() returns a snapshot, not a delegate. See the design doc for details.

[codecov]

Codecov Report

❌ Patch coverage is 49.33333% with 38 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.8%. Comparing base (dfdc478) to head (36d0e77).

Files with missing lines	Patch %	Lines
opentelemetry-sdk/src/logs/batch_log_processor.rs	49.3%	38 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #3514     +/-   ##
=======================================
- Coverage   82.8%   82.8%   -0.1%     
=======================================
  Files        130     130             
  Lines      27289   27364     +75     
=======================================
+ Hits       22622   22680     +58     
- Misses      4667    4684     +17     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.