fix(render): convert Trace::width through the UI scale, and fold the controller's tolerance conversion into Style::device()

docs/superpowers/plans/2026-09-28-plot-dpi-grid-follow.md

@@ -19,7 +19,7 @@
 - **Data-space geometry never scales.** Only chrome does. `dense_threshold_spp` stays a *device*-pixel ratio (it governs rasterisation density, not physical size); everything a user perceives as a size — font px, stroke widths, paddings, swatch and marker sizes, tick spacing — is logical.
 - The renderer clears only its viewport region via scissor and hands back host GL state; `renderer.hpp`'s documented residue list stays accurate.
 - OpenGL 3.3 core baseline.
-- The 156 existing test cases keep passing; any signature change is called out in the task report.
+- The 164 existing test cases keep passing (they are 171 from Task 1 on, which adds 7 of its own); any signature change is called out in the task report.
 - Every task ends with `cmake --preset linux-debug && cmake --build build/debug && ctest --test-dir build/debug --output-on-failure` green and warning-clean, and commits with the message in its own section.
 
 ---
@@ -74,7 +74,7 @@ git add -A && git commit -m "feat(render): logical style metrics with a UI scale
 - Test: `tests/test_plot.cpp`, `tests/test_render_smoke.cpp`
 
 **Interfaces:**
-- Consumes: the tick set (`plot::Tick{f64 value, bool major, std::string label}` and `generate_ticks`'s `include_minors`), `Style` from Task A.
+- Consumes: the tick set (`plot::Tick{f64 value, bool major, std::string label}` and `generate_ticks`'s `include_minors`), `Style` from Task 1 (`ui_scale`, `device()`, and the logical/device classification it wrote down, which is what makes the two new widths logical).
 - Produces: `Style::grid_major_width`, `Style::grid_minor_width` (logical), `Style::grid_major_visible`, `Style::grid_minor_visible`; grid segments tagged with their class.
 
 - [ ] **Step 1: Per-class style**