btSoftBodySolver_OpenCL::setDefaultWorkgroupSize to customize the work group size.

Thanks to Simon Green for the feedback, see also Issue 419

Added BT_PROFILE for "predictUnconstraintMotionSoftBody"

Added a few missing destructors.
Added AllMemoryBarrier
Thanks to Lee Howes for the commit in the branch.

src/BulletMultiThreaded/GpuSoftBodySolvers/OpenCL/btSoftBodySolver_OpenCL.h

@@ -185,6 +185,7 @@ private:
 
 	bool m_shadersInitialized;
 
+
 	/** 
 	 * Cloths owned by this solver.
 	 * Only our cloths are in this array.
@@ -246,6 +247,8 @@ private:
 	cl_command_queue	m_cqCommandQue;
 	cl_context			m_cxMainContext;
 
+	size_t				m_defaultWorkGroupSize;
+
 
 	/**
 	 * Compile a compute shader kernel from a string and return the appropriate cl_kernel object.
@@ -327,6 +330,15 @@ public:
 	virtual void predictMotion( float solverdt );
 
 	virtual void copySoftBodyToVertexBuffer( const btSoftBody *const softBody, btVertexBufferDescriptor *vertexBuffer );
+
+	virtual void	setDefaultWorkgroupSize(size_t workGroupSize)
+	{
+		m_defaultWorkGroupSize = workGroupSize;
+	}
+	virtual size_t	getDefaultWorkGroupSize() const
+	{
+		return m_defaultWorkGroupSize;
+	}
 }; // btOpenCLSoftBodySolver
 
 #endif // #ifndef BT_SOFT_BODY_SOLVER_OPENCL_H