Update imports

Remove dead code and add GPU VRAM profiling

Update src/alignment/special_segments.py

Replace midpoint split with MFA-based splitting for combined/fused segments

Eliminate in-decorator CPU fallback to prevent CUDA state poisoning

Stop CUDA state poisoning by not invalidating model caches during quota fallback

Add debug logging

Prevent CPU fallback from poisoning CUDA state via GPU model parking

Update special segments

Add tasleem special segment

Skip false missing-words flags across chapter boundaries and document special_type API field

Skip false missing-words flags after Surah 1 global reanchor and detect transitions at chapter boundaries

Add transition segment detection for non-Quranic phrases (Amin, Takbir, Tahmeed)

Add GPU retry for SDK worker errors instead of immediate CPU fallback

Fix GPU cross-user contamination: scope errors per-request, not process-wide

Modified anchor strategy to not bias towards longer surahs

Eliminate CUDA state poisoning at root cause, not just detection

Fix CUDA state poisoning by eliminating CUDA ops outside GPU leases

Prevent CUDA state poisoning and add thread-safe GPU lifecycle management

Harden GPU fallback to prevent CUDA init from permanently poisoning process

Restructure src/ for cleaner modularity

Flag missing words after cross-chapter Basmala transitions

Initial commit