FiberHMM 2.15.0 — bigBed Sample-tag fix + fix-bigbed + faster extract sort

• Fix: multi-layer bigBed tracks could lose layers in FiberBrowser when split/genotype-filtered pools were loaded side by side. The autoSQL Sample: tag is now sanitized to a dot/space-free, pool-distinct token so sibling pools no longer collapse to the same tag.
• New: fiberhmm-utils fix-bigbed repairs the embedded Sample: tag in existing bigBeds (rebuild via bigBedToBed → bedToBigBed, no re-extract).
• Faster fiberhmm-extract sort: runs under LC_ALL=C and adds -S/--sort-mem and --sort-parallel.