Describe the estimate's actual blind spots in the tower comments

The estimate no longer reports a tower for a single filament whose flush matrix
purges, so the comments that justify calling prime_tower_is_printed() instead of
reading a depth now cite what still holds: it reads neither enable_prime_tower
nor print_sequence. GCodeViewer's comments name render_scene(), the function
that replaced the render() they still pointed at, and the pass contract mentions
the toolhead boxes it draws.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

src/libslic3r/GCode/WipeTowerEstimate.hpp

@@ -45,8 +45,8 @@ WipeTowerFootprint estimate_wipe_tower_footprint(const ConfigBase
                                                  double                           max_object_height);
 
 // Whether a prime tower is PRINTED for a plate. The footprint estimate above answers how big a
-// tower is and reports one for a single filament whenever the flush matrix purges, so it cannot
-// answer this; the authority is DynamicPrintConfig::normalize_fdm_2(), which clears
+// tower is and never reads enable_prime_tower or print_sequence, so it reports a size for a plate
+// that prints no tower at all; the authority is DynamicPrintConfig::normalize_fdm_2(), which clears
 // enable_prime_tower before the plate is sliced. This mirrors that rule so pre-slice consumers can
 // ask it without mutating a config, and a test pins the two together.
 //

src/slic3r/GUI/GCodeViewer.cpp

@@ -1633,7 +1633,7 @@ void GCodeViewer::reset()
 // IDEX/IQEX: resolve where every secondary carriage marker sits, relative to the primary,
 // for one (printer preset, active mode, plate bed) triple. This parses the mode string
 // several times over and rebuilds the zone grid, and every one of its inputs is preset or
-// plate state, so render() runs it only when GCodeViewer::ImexMarkerKey changes and replays
+// plate state, so render_scene() runs it only when GCodeViewer::ImexMarkerKey changes and replays
 // the returned plan on all the frames in between.
 //
 // `mode` is already resolved (the plate's mode beats the process preset). An empty

src/slic3r/GUI/GCodeViewer.hpp

@@ -214,7 +214,7 @@ private:
     bool m_fold = {false};
     std::string m_marker_filename;      // cached for lazy secondary marker init
 
-    // IDEX/IQEX: everything render() needs to place the secondary carriage markers and the
+    // IDEX/IQEX: everything render_scene() needs to place the secondary carriage markers and the
     // toolhead footprint boxes, resolved from the printer preset, the active mode and the
     // plate bed. Resolving it walks the mode string through compute_imex_zone_layout()
     // (several string parses plus a zone-grid rebuild), and none of its inputs change
@@ -365,7 +365,7 @@ public:
     //BBS: add all plates filament statistics
     void render_all_plates_stats(const std::vector<const GCodeProcessorResult*>& gcode_result_list, bool show = true) const;
     //BBS: GUI refactor: add canvas width and height
-    // Shells, toolpaths and the sequential marker, drawn in 3D.
+    // Shells, toolpaths, the sequential markers and the IDEX/IQEX toolhead boxes, drawn in 3D.
     void render_scene(int canvas_width, int canvas_height);
     // Legend, sliders, the marker's position window and the G-code window, all ImGui.
     void render_overlay(int canvas_width, int canvas_height, int right_margin);

src/slic3r/GUI/PartPlate.cpp

@@ -1122,9 +1122,9 @@ Polygon PartPlate::imex_wipe_tower_hull() const
     // Whether a tower is PRINTED is normalize_fdm_2's rule, shared with the slicer through
     // prime_tower_is_printed() rather than re-derived here - re-deriving it is what let this
     // validate a tower the slicer had already cancelled. Do not substitute the footprint
-    // estimate: it reports a tower for a single filament whenever the flush matrix purges
-    // (SEMM + purge_in_prime_tower, the Klipper default), which hard-blocks a plate with
-    // nothing drawn on screen to move.
+    // estimate: it never reads enable_prime_tower or print_sequence, so it still sizes a tower
+    // for a plate that switched the tower off or prints by object, which would hard-block a
+    // plate with nothing drawn on screen to move.
     //
     // Counts are the ones normalize_fdm_2 is handed: filament slots as authored, so a mixed
     // slot counts once, and distinct objects rather than instances. filament_is_mixed is a