feat(isolation): DynamicVRAM compatibility for process isolation

DynamicVRAM's on-demand model loading/offloading conflicted with  process isolation in three ways: RPC tensor transport stalls from mid-call GPU offload, race conditions between model lifecycle and active RPC operations, and false positive memory leak detection from changed finalizer patterns.

- Marshal CUDA tensors to CPU before RPC transport for dynamic models
- Add operation state tracking + quiescence waits at workflow boundaries
- Distinguish proxy reference release from actual leaks in cleanup_models_gc
- Fix init order: DynamicVRAM must initialize before isolation proxies
- Add RPC timeouts to prevent indefinite hangs on model unavailability
- Prevent proxy-of-proxy chains from DynamicVRAM model reload cycles
- Add torch.device/torch.dtype serializers for new DynamicVRAM RPC paths
- Guard isolation overhead so non-isolated workflows are unaffected
- Migrate env var to PYISOLATE_CHILD

tests-unit/app_test/frontend_manager_test.py

@@ -49,6 +49,12 @@ def mock_provider(mock_releases):
     return provider
 
 
+@pytest.fixture(autouse=True)
+def clear_cache():
+    import utils.install_util
+    utils.install_util.PACKAGE_VERSIONS = {}
+
+
 def test_get_release(mock_provider, mock_releases):
     version = "1.0.0"
     release = mock_provider.get_release(version)