Add README files for examples and benchmark_results

- Add examples/README.md with usage instructions for all examples
- Add benchmark_results/README.md with benchmark usage guide
- Add reactor_owngil_example.erl and reactor_subinterp_example.erl
- Remove old benchmark error logs

Author: benoitc

benchmark_results/README.md

@@ -0,0 +1,93 @@
+# Benchmark Results
+
+This directory stores benchmark results for tracking performance across versions.
+
+## Running Benchmarks
+
+All benchmark scripts are in the `examples/` directory. Compile first with `rebar3 compile`.
+
+### Quick Benchmark
+
+```bash
+escript examples/benchmark.erl --quick
+```
+
+### Full Benchmark
+
+```bash
+escript examples/benchmark.erl --full
+```
+
+### Version Comparison Benchmark
+
+```bash
+escript examples/benchmark_compare.erl
+```
+
+This outputs a summary table useful for comparing performance between versions.
+
+### Concurrency Benchmark
+
+```bash
+escript examples/benchmark.erl --concurrent
+```
+
+### Channel API Benchmark
+
+```bash
+escript examples/bench_channel.erl
+```
+
+Compares Channel API throughput at various message sizes.
+
+### Reactor Modes Benchmark
+
+```bash
+escript examples/bench_reactor_modes.erl
+```
+
+Compares Reactor performance in worker mode vs subinterpreter mode (Python 3.12+).
+
+### Event Loop Benchmark
+
+```bash
+rebar3 shell
+> run_benchmark:run().
+```
+
+Compares Erlang event loop vs standard asyncio event loop performance.
+
+## Saving Results
+
+To save benchmark results for comparison:
+
+```bash
+# Run and save to timestamped file
+escript examples/benchmark_compare.erl > benchmark_results/$(date +%Y%m%d_%H%M%S).txt
+```
+
+## Result Files
+
+Result files are named with timestamps: `YYYYMMDD_HHMMSS.txt`
+
+Example files:
+- `reactor_modes_20260309_093558.txt` - Reactor modes comparison results
+
+## Key Metrics
+
+When comparing versions, focus on:
+
+1. **Sync call latency** - Time per `py:call()` operation
+2. **Cast throughput** - Non-blocking `py:cast()` operations per second
+3. **Concurrent throughput** - Performance under parallel load
+4. **Channel throughput** - Messages per second at various sizes
+5. **Reactor modes** - Worker vs subinterpreter performance
+
+## System Information
+
+Benchmarks automatically print system info including:
+- Erlang/OTP version
+- Number of schedulers
+- Python version
+- Execution mode (free_threaded, subinterp, multi_executor)
+- Max concurrent operations

benchmark_results/baseline_20260224_133948.txt.log

@@ -0,0 +1,169 @@
+# Examples
+
+This directory contains runnable examples demonstrating erlang_python features.
+
+## Prerequisites
+
+```bash
+rebar3 compile
+```
+
+## Basic Examples
+
+### basic_example.erl
+Basic Python calls, eval, and async operations.
+```bash
+escript examples/basic_example.erl
+```
+
+### streaming_example.erl
+Streaming from Python generators.
+```bash
+escript examples/streaming_example.erl
+```
+
+### shared_state_example.erl
+Shared state between Erlang and Python workers.
+```bash
+escript examples/shared_state_example.erl
+```
+
+### logging_example.erl
+Python logging forwarded to Erlang logger, plus distributed tracing.
+```bash
+escript examples/logging_example.erl
+```
+
+### erlang_concurrency.erl
+Demonstrates parallel execution using BEAM processes from Python.
+```bash
+escript examples/erlang_concurrency.erl
+```
+
+### reentrant_callback.erl / reentrant_demo.erl
+Python calling Erlang calling Python (reentrant callbacks).
+```bash
+escript examples/reentrant_callback.erl
+escript examples/reentrant_demo.erl
+```
+
+## AI/ML Examples
+
+These require additional Python packages.
+
+### semantic_search.erl
+Semantic search with sentence-transformers.
+```bash
+python3 -m venv /tmp/ai-venv
+/tmp/ai-venv/bin/pip install sentence-transformers numpy
+escript examples/semantic_search.erl
+```
+
+### rag_example.erl
+Retrieval-Augmented Generation with Ollama.
+```bash
+/tmp/ai-venv/bin/pip install sentence-transformers numpy requests
+ollama pull llama3.2
+escript examples/rag_example.erl
+```
+
+### ai_chat.erl
+Interactive AI chat.
+```bash
+escript examples/ai_chat.erl
+```
+
+### embedder_example.erl
+Text embeddings generation.
+```bash
+escript examples/embedder_example.erl
+```
+
+## Reactor Examples
+
+### reactor_echo.erl
+Simple echo server using Reactor API.
+```bash
+escript examples/reactor_echo.erl
+```
+
+### reactor_subinterp_example.erl
+Reactor with subinterpreter isolation (Python 3.12+).
+```bash
+escript examples/reactor_subinterp_example.erl
+```
+
+### reactor_owngil_example.erl
+Reactor with OWN_GIL subinterpreters (Python 3.12+).
+```bash
+escript examples/reactor_owngil_example.erl
+```
+
+## Benchmarks
+
+### benchmark.erl
+General performance benchmark with options.
+```bash
+escript examples/benchmark.erl --quick    # Quick run
+escript examples/benchmark.erl --full     # Full benchmark
+escript examples/benchmark.erl --concurrent  # Concurrency focus
+```
+
+### benchmark_compare.erl
+Version comparison benchmark with summary table.
+```bash
+escript examples/benchmark_compare.erl
+```
+
+### bench_channel.erl
+Channel API throughput benchmark.
+```bash
+escript examples/bench_channel.erl
+```
+
+### bench_reactor_modes.erl
+Reactor worker vs subinterpreter benchmark.
+```bash
+escript examples/bench_reactor_modes.erl
+```
+
+### bench_event_loop.erl
+Event loop performance comparison.
+```bash
+escript examples/bench_event_loop.erl
+```
+
+### bench_reactor_buffer.erl
+Reactor buffer performance.
+```bash
+escript examples/bench_reactor_buffer.erl
+```
+
+### bench_resource_pool.erl
+Resource pool benchmark.
+```bash
+escript examples/bench_resource_pool.erl
+```
+
+### run_benchmark.erl
+Event loop benchmark (run in shell).
+```bash
+rebar3 shell
+> run_benchmark:run().
+```
+
+## Elixir
+
+### elixir_example.exs
+Elixir integration example.
+```bash
+elixir --erl "-pa _build/default/lib/erlang_python/ebin" examples/elixir_example.exs
+```
+
+## Other
+
+### gen_test.erl
+Test generator for development.
+```bash
+escript examples/gen_test.erl
+```