Fixing Multiple Profiler Issues:

1- Fixing Output Buffer Issues
2- Metrics Instances created for all GPUs in Init
3- Multi CPU/GPU/KernArg Pools are initialized for every Agent in Init
4- Lowering OverHead in the Packet WriteInterceptor in the ProxyQueue for both Profile Counting & ATT

Change-Id: Ic3d78156af8405bb134d01584574c339237d265f


[ROCm/rocprofiler commit: 00ecca25c7]

projects/rocprofiler/src/util/hsa_rsrc_factory.cpp

@@ -489,7 +489,7 @@ uint8_t* HsaRsrcFactory::AllocateLocalMemory(const AgentInfo* agent_info, size_t
 }
 
 // Allocate memory to pass kernel parameters.
-// Memory is alocated accessible for all CPU agents and for GPU given by AgentInfo parameter.
+// Memory is allocated accessible for all CPU agents and for GPU given by AgentInfo parameter.
 // @param agent_info Agent from whose memory region to allocate
 // @param size Size of memory in terms of bytes
 // @return uint8_t* Pointer to buffer, null if allocation fails.