Rider import benchmark

The rider_import_benchmark test exercises the rider importer against a large synthetic rider file to monitor execution time and memory use. It can be run manually or through CTest and is intended to provide a reproducible measurement before and after rider-related optimizations.

Files involved

tests/data/rider_large.txt – 1,100+ fixture entries distributed across multiple hangs and a floor section.
tests/rider_import_benchmark.cpp – small console benchmark that measures import duration and process memory usage while reporting the resulting fixture and truss counts.

Building and running

Configure the test project (requires the same dependencies as the other C++ tests, including wxWidgets and tinyxml2):
```
cmake -S tests -B build/tests
```

Build the benchmark target:

cmake --build build/tests --target rider_import_benchmark

Run the benchmark with the default large rider file (the second argument is the iteration count):
```
./build/tests/rider_import_benchmark tests/data/rider_large.txt 3
```
Or execute it through CTest:
```
cd build/tests
ctest -R RiderImportBenchmark
```

Comparing results

Record the source revision, compiler and build type, host hardware, iteration count, elapsed time, peak RSS, and imported object counts with each result. Compare before and after runs on the same provisioned host; this document does not define a portable timing or memory threshold.