refactor: Reorganize tests into subsystem subdirectories

Restructured test suite for better organization and targeted testing:

**Structure:**
- src/tests/audio/    - 15 audio system tests
- src/tests/gpu/      - 12 GPU/shader tests
- src/tests/3d/       - 6 3D rendering tests
- src/tests/assets/   - 2 asset system tests
- src/tests/util/     - 3 utility tests
- src/tests/common/   - 3 shared test helpers
- src/tests/scripts/  - 2 bash test scripts (moved conceptually, not physically)

**CMake changes:**
- Updated add_demo_test macro to accept LABEL parameter
- Applied CTest labels to all 36 tests for subsystem filtering
- Updated all test file paths in CMakeLists.txt
- Fixed common helper paths (webgpu_test_fixture, etc.)
- Added custom targets for subsystem testing:
  - run_audio_tests, run_gpu_tests, run_3d_tests
  - run_assets_tests, run_util_tests, run_all_tests

**Include path updates:**
- Fixed relative includes in GPU tests to reference ../common/

**Documentation:**
- Updated doc/HOWTO.md with subsystem test commands
- Updated doc/CONTRIBUTING.md with new test organization
- Updated scripts/check_all.sh to reflect new structure

**Verification:**
- All 36 tests passing (100%)
- ctest -L <subsystem> filters work correctly
- make run_<subsystem>_tests targets functional
- scripts/check_all.sh passes

Backward compatible: make test and ctest continue to work unchanged.

handoff(Gemini): Test reorganization complete. 36/36 tests passing.

1 files changed, 215 insertions, 0 deletions

diff --git a/src/tests/audio/test_mock_backend.cc b/src/tests/audio/test_mock_backend.cc
new file mode 100644
index 0000000..defd73d
--- /dev/null
+++ b/src/tests/audio/test_mock_backend.cc
@@ -0,0 +1,215 @@
+// This file is part of the 64k demo project.
+// It tests the MockAudioBackend implementation.
+// Verifies event recording, time tracking, and synth integration.
+
+#include "audio/audio.h"
+#include "audio/backend/mock_audio_backend.h"
+#include "audio/synth.h"
+#include <assert.h>
+#include <cmath>
+#include <stdio.h>
+
+#if !defined(STRIP_ALL)
+
+void test_event_recording() {
+  MockAudioBackend backend;
+
+  // Initially no events
+  assert(backend.get_events().size() == 0);
+  assert(backend.get_current_time() == 0.0f);
+
+  // Simulate voice trigger
+  backend.on_voice_triggered(0.5f, 3, 0.75f, -0.25f);
+
+  // Verify event recorded
+  const auto& events = backend.get_events();
+  assert(events.size() == 1);
+  assert(events[0].timestamp_sec == 0.5f);
+  assert(events[0].spectrogram_id == 3);
+  assert(events[0].volume == 0.75f);
+  assert(events[0].pan == -0.25f);
+
+  // Record multiple events
+  backend.on_voice_triggered(1.0f, 5, 1.0f, 0.0f);
+  backend.on_voice_triggered(1.5f, 3, 0.5f, 0.5f);
+
+  assert(backend.get_events().size() == 3);
+  assert(events[1].timestamp_sec == 1.0f);
+  assert(events[2].timestamp_sec == 1.5f);
+
+  // Clear events
+  backend.clear_events();
+  assert(backend.get_events().size() == 0);
+
+  printf("Event recording test PASSED\n");
+}
+
+void test_time_tracking() {
+  MockAudioBackend backend;
+
+  // Test manual time advance
+  assert(backend.get_current_time() == 0.0f);
+
+  backend.advance_time(0.5f);
+  assert(backend.get_current_time() == 0.5f);
+
+  backend.advance_time(1.0f);
+  assert(backend.get_current_time() == 1.5f);
+
+  // Test time setting
+  backend.set_time(10.0f);
+  assert(backend.get_current_time() == 10.0f);
+
+  printf("Time tracking test PASSED\n");
+}
+
+void test_frame_rendering() {
+  MockAudioBackend backend;
+
+  // Simulate frame rendering (32000 Hz sample rate)
+  // 1 second = 32000 frames
+  backend.on_frames_rendered(16000); // 0.5 seconds
+  assert(std::abs(backend.get_current_time() - 0.5f) < 0.001f);
+
+  backend.on_frames_rendered(16000); // Another 0.5 seconds
+  assert(std::abs(backend.get_current_time() - 1.0f) < 0.001f);
+
+  backend.on_frames_rendered(32000); // 1 second
+  assert(std::abs(backend.get_current_time() - 2.0f) < 0.001f);
+
+  printf("Frame rendering test PASSED\n");
+}
+
+void test_synth_integration() {
+  MockAudioBackend backend;
+  audio_set_backend(&backend);
+
+  synth_init();
+
+  // Create dummy spectrogram
+  float data[DCT_SIZE * 10] = {0};
+  data[0] = 100.0f; // DC component
+
+  Spectrogram spec = {data, data, 10};
+  int spec_id = synth_register_spectrogram(&spec);
+  assert(spec_id >= 0);
+
+  // Trigger voice - should be recorded at time 0
+  synth_trigger_voice(spec_id, 0.8f, -0.3f);
+
+  // Verify event recorded
+  const auto& events = backend.get_events();
+  assert(events.size() == 1);
+  assert(events[0].timestamp_sec == 0.0f); // Before any rendering
+  assert(events[0].spectrogram_id == spec_id);
+  assert(events[0].volume == 0.8f);
+  assert(events[0].pan == -0.3f);
+
+  // Render some frames to advance time
+  float output[1024] = {0};
+  synth_render(output, 512); // ~0.016 sec at 32kHz
+
+  // Verify synth updated its time
+  // (Note: synth time is internal, mock doesn't track it from render)
+
+  // Trigger another voice after rendering
+  synth_trigger_voice(spec_id, 1.0f, 0.5f);
+
+  assert(events.size() == 2);
+  // Second trigger should have timestamp > 0
+  assert(events[1].timestamp_sec > 0.0f);
+  assert(events[1].timestamp_sec < 0.02f); // ~512 frames = ~0.016 sec
+
+  printf("Synth integration test PASSED\n");
+}
+
+void test_multiple_voices() {
+  MockAudioBackend backend;
+  audio_set_backend(&backend);
+
+  synth_init();
+
+  // Create multiple spectrograms
+  float data1[DCT_SIZE * 5] = {0};
+  float data2[DCT_SIZE * 5] = {0};
+  float data3[DCT_SIZE * 5] = {0};
+
+  Spectrogram spec1 = {data1, data1, 5};
+  Spectrogram spec2 = {data2, data2, 5};
+  Spectrogram spec3 = {data3, data3, 5};
+
+  int id1 = synth_register_spectrogram(&spec1);
+  int id2 = synth_register_spectrogram(&spec2);
+  int id3 = synth_register_spectrogram(&spec3);
+
+  // Trigger multiple voices at once
+  synth_trigger_voice(id1, 1.0f, -1.0f);
+  synth_trigger_voice(id2, 0.5f, 0.0f);
+  synth_trigger_voice(id3, 0.75f, 1.0f);
+
+  // Verify all recorded
+  const auto& events = backend.get_events();
+  assert(events.size() == 3);
+
+  // Verify each has correct properties
+  assert(events[0].spectrogram_id == id1);
+  assert(events[1].spectrogram_id == id2);
+  assert(events[2].spectrogram_id == id3);
+
+  assert(events[0].volume == 1.0f);
+  assert(events[1].volume == 0.5f);
+  assert(events[2].volume == 0.75f);
+
+  assert(events[0].pan == -1.0f);
+  assert(events[1].pan == 0.0f);
+  assert(events[2].pan == 1.0f);
+
+  printf("Multiple voices test PASSED\n");
+}
+
+void test_audio_render_silent_integration() {
+  MockAudioBackend backend;
+  audio_set_backend(&backend);
+
+  audio_init();
+  synth_init();
+
+  // Create a spectrogram
+  float data[DCT_SIZE * 5] = {0};
+  Spectrogram spec = {data, data, 5};
+  int spec_id = synth_register_spectrogram(&spec);
+
+  // Trigger at t=0
+  synth_trigger_voice(spec_id, 1.0f, 0.0f);
+
+  // Simulate 2 seconds of silent rendering (seek/fast-forward)
+  audio_render_silent(2.0f);
+
+  // Verify backend time advanced via on_frames_rendered
+  const float expected_time = 2.0f;
+  const float actual_time = backend.get_current_time();
+  assert(std::abs(actual_time - expected_time) < 0.01f); // 10ms tolerance
+
+  audio_shutdown();
+
+  printf("audio_render_silent integration test PASSED\n");
+}
+
+#endif /* !defined(STRIP_ALL) */
+
+int main() {
+#if !defined(STRIP_ALL)
+  printf("Running MockAudioBackend tests...\n");
+  test_event_recording();
+  test_time_tracking();
+  test_frame_rendering();
+  test_synth_integration();
+  test_multiple_voices();
+  test_audio_render_silent_integration();
+  printf("All MockAudioBackend tests PASSED\n");
+  return 0;
+#else
+  printf("MockAudioBackend tests skipped (STRIP_ALL enabled)\n");
+  return 0;
+#endif /* !defined(STRIP_ALL) */
+}