Add the possibility to read a user-defined event bank in CaloHits.d

I open this issue because frequently it happens to me to customize the NA64 MC code https://gitlab.cern.ch/P348/na64-simulation in order to produce, in CaloHits.d, one or more new banks for specific studies. For example, to study the decay of hadrons, I added a new section called DECAYTRUTH, to record the decay products of the primary particle (in case this decays in the setup).

DECAYTRUTH
<n_lines following this>
PID Px Py Pz Vx Vy Vz
PID Px Py Pz Vx Vy Vz
...
PID Px Py Pz Vx Vy Vz

Then, I modified my local copy of p348reco to parse this block:

 else if (mytag == "DECAYTRUTH"){
       int nD=0;
       int PID;
       double Dpx,Dpy,Dpz;
       double Dvx,Dvy,Dvz;

       inFile >> nD;
       for (int iD=0;iD<nD;iD++){
         inFile>>nD>>Dpx>>Dpy>>Dpz>>Dvx>>Dvy>>Dvz;
         //Do something with the obtained values
       }
     }

This approach has the clear drawback of breaking compatibility with the central repository - any time a modification is introduced in p348reco.h and I do a git pull, I get a conflict.

I'd suggest to introduce a more general approach to allow any user to add any new bank in the CaloHits.d and have p348reco.h compatible with it.

1. Define a new block USERBANK with the following syntax:

USERBANK
n
line-1
line-2
...
line-n

with n being the number of lines following n (ideally >0, but should also handle the case n=0).

2. Introduce a new member in the struct MCtruth, vector<string> userbank
3. Add a code like the following to read the new section of CaloHits.d

else if (mytag == "USERBANK") {
      int nL;
      string ss;
      inFile >> nL;
      inFile.ignore(1000, '\n'); //to be compatible with the use of getline below
      for (int ii = 0; ii < nL; ii++) {
        getline(inFile, ss);
        e.mc->userbank.push_back(ss);
      }
      continue; //to avoid the ignore instruction at the end of the "mytag" parsing
    }

At this point, in any reconstruction code based on p348reco it will be possible to properly parse the content of USERBANK.

@akorneev @bbantoob @kirsanov @rdusaev may I ask your feedback on this?