Free G-code result and preview memory on reset (#16065)

GCodeProcessorResult declares a copy assignment, which suppresses the
implicit
move assignment, so `*result = std::move(processor.extract_result())`
binds to
the copy and duplicates the whole moves array: 96 bytes per move,
measured at
147.7 MiB in one allocation for a 597k facet model at 0.08 mm, with the
source
staying alive until the export returns.

Add the move assignment. It assigns exactly the same 39 members in the
same
order as the copy, including the ones the copy deliberately leaves
alone, so a
target that carries its own filament maps and nozzle type keeps them. A
move
constructor is intentionally not added: the type holds a std::mutex, so
it is
neither copy nor move constructible today and a partial one would leave
ten
members uninitialised.

reset() now releases the two print sized vectors rather than clearing
them. Its
callers are the paths that discard a result, so the memory went back
only at the
next slice.

ViewerImpl::reset() does the same for the two vectors sized by the
print, which
is what keeps a discarded preview resident, and counts m_vertices_colors
in the
reported CPU memory, where it was missing.

src/libslic3r/GCode/GCodeProcessor.cpp

@@ -2602,12 +2602,13 @@ void GCodeProcessorResult::reset() {
     //BBS: add mutex for protection of gcode result
     lock();
 
-    moves.clear();
+    // moves and lines_ends grow with the print: free them, as no reset result reuses the capacity.
+    clear_and_shrink(moves);
     plate_mass = {};
     object_masses.clear();
     body_masses.clear();
     support_masses.clear();
-    lines_ends.clear();
+    clear_and_shrink(lines_ends);
     printable_area = Pointfs();
     //BBS: add bed exclude area
     bed_exclude_area = Pointfs();

src/libvgcode/src/ViewerImpl.cpp

@@ -927,15 +927,17 @@ void ViewerImpl::reset()
     m_used_extruders.clear();
     m_total_time = { 0.0f, 0.0f };
     m_travels_time = { 0.0f, 0.0f };
-    m_vertices.clear();
-    m_vertices_colors.clear();
     // swap rather than clear: these are sized by the print, and a reset means the memory
     // should go back, not sit reserved until the next load
+    std::vector<PathVertex>().swap(m_vertices);
+    std::vector<float>().swap(m_vertices_colors);
     for (std::vector<float>& times : m_layer_start_times)
         std::vector<float>().swap(times);
     std::vector<uint32_t>().swap(m_layer_first_vertex);
     m_layers_in_vertex_order = false;
-    m_valid_lines_bitset.clear();
+    // BitSet::clear() only zeroes the bits, it keeps the blocks allocated; load() builds a new
+    // bitset anyway and it is never read while m_vertices is empty
+    m_valid_lines_bitset = BitSet<>();
 #if VGCODE_ENABLE_COG_AND_TOOL_MARKERS
     m_cog_marker.reset();
 #endif // VGCODE_ENABLE_COG_AND_TOOL_MARKERS
@@ -1834,6 +1836,7 @@ size_t ViewerImpl::get_used_cpu_memory() const
     ret += sizeof(m_extrusion_roles_colors);
     ret += sizeof(m_options_colors);
     ret += STDVEC_MEMSIZE(m_vertices, PathVertex);
+    ret += STDVEC_MEMSIZE(m_vertices_colors, float);
     for (const std::vector<float>& times : m_layer_start_times)
         ret += STDVEC_MEMSIZE(times, float);
     ret += STDVEC_MEMSIZE(m_layer_first_vertex, uint32_t);