remove readback hacks, allows to read/write guest memory directly

src/graphics/host_gpu/renderer/gpuResourceManager.h

@@ -9,7 +9,6 @@
 #include "graphics/host_gpu/renderer/textureCache.h"
 
 #include <cstdint>
-#include <mutex>
 
 namespace Libs::Graphics {
 
@@ -26,14 +25,6 @@ public:
 
 	[[nodiscard]] bool HandleFault(PageFaultAccess access, uint64_t fault_vaddr) noexcept;
 	[[nodiscard]] bool IsMapped(uint64_t vaddr, uint64_t size) const noexcept;
-	// Serializes synchronous command-processor guest-memory access against buffer/image ownership
-	// publication. Address metadata keeps this seam ready for a future page-range lock.
-	[[nodiscard]] std::unique_lock<ResourceMutex>
-	LockCommandProcessorGuestAccess(uint64_t vaddr, uint64_t size);
-	// Keeps an asynchronous host callback's validation and guest write in one cache transaction.
-	// Cached buffer/image destinations are unsupported and fail before the callback can fault.
-	[[nodiscard]] std::unique_lock<ResourceMutex> LockAsynchronousGuestWrite(uint64_t vaddr,
-	                                                                        uint64_t size);
 	void               MapMemory(uint64_t vaddr, uint64_t size, GpuAccess access);
 	void               UnmapMemory(uint64_t vaddr, uint64_t size, GpuAccess access);
 	void FillBuffer(CommandBuffer* command, uint64_t vaddr, uint64_t size, uint32_t value);
@@ -48,7 +39,7 @@ private:
 	PageManager   m_page_manager;
 	ResourceMutex m_resource_mutex;
 	BufferCache   m_buffer_cache;
-	TextureCache m_texture_cache;
+	TextureCache  m_texture_cache;
 };
 
 } // namespace Libs::Graphics