memory_manager: Use GPUVAdddr, not PAddr, for GPU addresses.

src/video_core/command_processor.cpp

@@ -90,9 +90,7 @@ void GPU::WriteReg(u32 method, u32 subchannel, u32 value, u32 remaining_params)
 }
 
 void GPU::ProcessCommandList(GPUVAddr address, u32 size) {
-    // TODO(Subv): PhysicalToVirtualAddress is a misnomer, it converts a GPU VAddr into an
-    // application VAddr.
-    const VAddr head_address = memory_manager->PhysicalToVirtualAddress(address);
+    const VAddr head_address = memory_manager->GpuToCpuAddress(address);
     VAddr current_addr = head_address;
     while (current_addr < head_address + size * sizeof(CommandHeader)) {
         const CommandHeader header = {Memory::Read32(current_addr)};