Rust CSV Parsing Engine

Tried to maximize csv ingestion and parsing speed, finally ended up with the following

1. Memory Mapping: File is memory-mapped (mmap) and partitioned into equal chunks, using memchr to align thread boundaries to newlines.
2. Zero Allocation Scanning: Standard csv crate is bypassed in favor of a custom stack allocated byte scanner that uses memchr for SIMD comma splitting. Can switch to csv crate too if data consists of complex columns (e.g. commas inside quotes)
3. Fast Float Parsing: Fast-float crate parses numbers directly from raw &[u8] slices.

Performance Results

Hardware: HP Omen 16 (16 Threads)

Metric	Result
Test File	ohlcv/test.csv (4096 MB)
Concurrency	16 Threads
Rows Processed	76,820,511
Matching Rows	33,824,057
Total Time	1.16s
Throughput	3523.24 MB/s