Re #9543. Handle the addition of monitor workspace chunks.

This step explicitly adds monitor workspaces in addition to the main
data. This means that if the monitor workspace is still present after
the processing step (a user could extract it there if they wanted), it
will be available for the post-processing or be carried right through
to the output if there is no post-processing step.

Code/Mantid/Framework/LiveData/src/LoadLiveData.cpp

@@ -286,6 +286,18 @@ namespace LiveData
    */
   void LoadLiveData::addMatrixWSChunk(const std::string& algoName, Workspace_sptr accumWS, Workspace_sptr chunkWS)
   {
+    // Handle the addition of the internal monitor workspace, if present
+    auto accumMW = boost::dynamic_pointer_cast<MatrixWorkspace>(accumWS);
+    auto chunkMW = boost::dynamic_pointer_cast<MatrixWorkspace>(chunkWS);
+    if ( accumMW && chunkMW )
+    {
+      auto accumMon = accumMW->monitorWorkspace();
+      auto chunkMon = chunkMW->monitorWorkspace();
+
+      if ( accumMon && chunkMon ) accumMon += chunkMon;
+    }
+
+    // Now do the main workspace
     IAlgorithm_sptr alg = this->createChildAlgorithm(algoName);
     alg->setProperty("LHSWorkspace", accumWS);
     alg->setProperty("RHSWorkspace", chunkWS);
@@ -297,9 +309,8 @@ namespace LiveData
     }
     else
     {
-      // Is this really necessary?
-
       // Get the output as the generic Workspace type
+      // This step is necessary for when we are operating on MD workspaces (PlusMD)
       Property * prop = alg->getProperty("OutputWorkspace");
       IWorkspaceProperty * wsProp = dynamic_cast<IWorkspaceProperty*>(prop);
       if (!wsProp)