docs: Add immediate TODOs for R=4 motion mask, weak ranging, and infinity target filtering

docs/project-details.md

@@ -558,6 +558,10 @@ This section captures the latest validated state before pausing Global LMA work
 
 ### Next TODO (priority order)
 1. Performance pass: identify current bottlenecks and low-hanging optimizations.
+   - **Two-stage motion vector calculation:** Repeat correlation with a tight mask (R=4) around located targets to significantly decrease the noise integration area and eliminate motion blur.
+   - **Fix ranging:** Correct ranging logic for weak targets.
+   - **Radar Annotation Cleanup:** Implement the GUI checkbox logic to hide/ignore false-positive targets at infinity (atmospheric turbulence) in the video output, *except* for the ground-truth flight log target.
+   - **Flight Log Robustness:** Test Latvia datasets that lack flight logs to ensure the pipeline degrades gracefully without crashing.
 2. **Agent-Driven Algorithm Evolution (Future Roadmap):**
    - **Adaptive Integration Windows:** Implement a competitive selection mechanism (e.g., 1x vs 4x integration) where the pipeline automatically selects the motion vector that yields the highest Contrast-to-Noise Ratio (CNR), rather than relying on hardcoded thresholds.
    - **Fat Zero Auto-Scaling:** Automatically scale `cuas_rng_fz` (fat zero) down proportionally to the length of the temporal integration (longer integrations lower the stochastic noise floor, allowing closer-to-pure phase correlation).