Add local benchmark mode and Docker support

[lsk567]

Summary

This PR adds support for running timing benchmarks locally without connecting to a remote embedded device, along with Docker support for reproducible environments.

Changes

Benchmark Scripts:

• Add --local (-l) flag to run_benchmark.py to run benchmarks locally
• Add local execution functions: local_compile_cmake_project(), local_run_program(), local_run_trace_conversion()
• Add --local and --select flags to experiment_timing.py
• Make credentials.py import conditional (only required for remote mode)

Docker Support:

• Add Dockerfile with all dependencies (Java 17, Python, CMake, RISC-V toolchain, Verilator, etc.)
• Add docker-compose.yml for convenience

CI:

• Add GitHub Actions workflow for local benchmark testing

Test plan

• Tested run_benchmark.py --local --select=PingPong locally
• Tested experiment_timing.py --local --select=PingPong locally (all 3 schedulers: NP, LB, EGS)
• Verified CSV trace files and plots are generated correctly
• CI workflow runs successfully

Usage

# Run single benchmark locally
cd benchmarks
python3 scripts/run_benchmark.py --local --src=src --src-gen=src-gen --select=PingPong

# Run full timing experiment locally
python3 scripts/experiment_timing.py --local --select=PingPong

# Run in Docker
docker build -t emsoft25:local .
docker run -it -v $(pwd):/workspace emsoft25:local
source /setup_env.sh
cd /workspace/lingua-franca && ./gradlew assemble
cd /workspace/benchmarks && python3 scripts/experiment_timing.py --local

🤖 Generated with Claude Code

[lsk567]

@copilot please review this PR

[Copilot]

@lsk567 I've opened a new pull request, #7, to work on those changes. Once the pull request is ready, I'll request review from you.