Add C++17 PERC Runtime Optimizer

[sethupathib]

Summary

Adds a strictly C++17 educational project for understanding PERC (Programmable Electrical Rules Checking) in physical design and experimenting with runtime optimizations.

White paper

Detailed PDF now in-repo: PERC Runtime Optimizer/docs/PERC_Runtime_Optimizer_Whitepaper.pdf (10 pages). Covers industrial PERC flow, why P2P/CD + extract/fill dominate runtime, the synthetic non-GDS data model, engines, and measured results. Regenerate with python3 docs/generate_whitepaper.py.

What PERC is (see also docs/PERC_OVERVIEW.md)

• Reliability checks beyond DRC/LVS: ESD clamps, floating gates, P2P resistance, current-density-style path checks
• Why it is slow: huge netlists, many pad→rail queries, full re-runs after ECO; industrially often extract/fill-bound

What the code does

Synthetic hierarchical netlist + resistor graph (Dijkstra P2P), then engines:

Mode	Optimization
baseline	Full-chip every time
roi	ESD-relevant device pruning
hierarchical	Per-block FG with shared index
parallel	Multi-threaded pad-pair P2P/CD
incremental / optimized	Metadata cache + ECO untouched-block reuse

Build / test

cd "PERC Runtime Optimizer"
make -j && make test
./perc_optimizer --bench --eco

Bench highlights: warm incremental ≫ cold; ECO recheck only touches changed blocks; parallel helps P2P-heavy runs.