Refinement: density records sent from pinned memory, and the L-BFGS direction in fused sweeps

Two byte-identical cuts to the per-evaluation host chain, which on a large model is twice the GPU's
work per evaluation (34 ms against 16 ms of kernels per weight-trial slot, measured in nsys) and so bounds
the refinement on one card with three slots, and the whole of it on a card with one slot.

- The gather's per-atom density records (38 MB on a 242k-atom model, sent on every B-factor evaluation)
  live in a PinnedVector and go up by DMA; the boxes go through a pinned staging buffer of the density
  object. ModelDensityGPU::SetAtoms and RefinementGPUEngine::SetDensities take a pointer and a count.
- RefinementMinimiser::Direction takes each pair's dot product in the same sweep that applies the previous
  pair's update to q: the dot products are serial chains of n additions, and the update's memory traffic
  now hides under them (20 -> 13 ms per call at 365k parameters). The same operations per element in the
  same order; the loop-2 coefficient stays written as (a[i] - b) with b on its own, because a[i] - rho * dot
  contracts into a fused multiply-add and changes the bits (checked bitwise in a standalone harness).
- A debug line with the GPU engine's bytes and the budget.

4nwv (open arm, 242k atoms), live flags, one lock hold, loaded box: refinement 42.35 -> 38.40 s;
refine.gpu.upload 8.14 -> 3.94 s. All outputs identical but provenance lines.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015CeEGgpKdfkCTJb3YNMJcu

image_analysis/structure_refinement/RefinementGPU.cpp

@@ -304,7 +304,7 @@ private:
         stage.reset();
         const RefinementStage upload("gpu.upload");
         if (std::find(changed.begin(), changed.end(), 1) != changed.end())
-            engine_->SetDensities(atoms_);
+            engine_->SetDensities(atoms_.data(), atoms_.size());
         engine_->SetPositions(pos);
     }
 
@@ -320,7 +320,7 @@ private:
     std::unique_ptr<RefinementGPUEngine> engine_;
     std::vector<const gemmi::Atom *> atom_ptrs_;   // the model's atoms at the last SetAtoms()
     std::vector<float> last_b_;
-    std::vector<ModelDensityAtom> atoms_;
+    PinnedVector<ModelDensityAtom> atoms_;   // pinned: a B rung sends them every evaluation (37 MB on a large model)
 };
 
 }  // namespace