Add CUDA GPU benchmark results

[apoorvalal]

Summary

• configure the benchmark environment for the CUDA GPU stack with torch, JAX, and CuPy support
• add torch-cuda, JAX, and CuPy benchmark result CSVs for base, AKM sweep, and AKM occupation benchmarks
• refresh benchmark figures and update plotting styles so full backend legends have distinct color/marker pairs

Verification

• pixi run -e benchmark benchmark
• pixi run -e benchmark benchmark-akm
• pixi run -e benchmark benchmark-akm-occupation
• PYFIXEST_BENCHMARK_INCLUDE_CUPY=1 full benchmark suite exited 0 for all three tasks
• pixi run pytest tests/test_modular_plotting.py

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Flag	Coverage Δ
core-tests	72.83% <ø> (ø)
test-r-core	51.28% <ø> (ø)
test-r-extended	19.00% <ø> (ø)
test-r-fixest	40.56% <ø> (ø)
tests-extended	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
pyfixest/utils/dgps.py	55.80% <ø> (ø)

... and 6 files with indirect coverage changes

🚀 New features to boost your workflow:

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