print out the 95% CL limits directly from the cpv plotting script.

It takes time to rerun the cpv limit script to get the log file, so it's better to print out the 95% CL limits directly from the plotting script. Here's how I did it:

// getting the 0.68 CL plot from input file.
  TString peakName="Posterior/PeakPlusMinusSigma";
  peakName += postName + "_limit";
  cout<< "first time " << peakName.Data()<< endl;

or

// getting the 0.95 CL plot from input file.
  TString peakName="Posterior/PeakPlusMinusSigma";
  peakName += postName;
  cout<< "2nd time " << peakName.Data()<< endl;

then
// get the stored values from the vector.
  postVals = *((TVectorD*) infile->Get(peakName.Data()));
  postPeak=postVals[0];
  postSigmaPlus=postVals[2];
  postSigmaMinus=postVals[1];
  postLimit=postVals[3];

  // printing the values
  cout<<"######CL 0.95 is from " << postSigmaMinus << " to " << postSigmaPlus << endl;


For example for tbtqb observed:

first time Posterior/PeakPlusMinusSigma_asymmetry_95

######CL 0.68 is from -0.5756 to 0.0714

2nd time Posterior/PeakPlusMinusSigma_asymmetry

posterior peak is at -0.23036 +0.15448 -0.16296, limit 0

######CL 0.95 is from -0.39332 to -0.07588

However it's a little confusing. I tried using
PeakPlusMinusSigma_asymmetry_limit this actually gives 68% CL.
I tried 2nd time using PeakPlusMinusSigma_asymmetry, this time it gives 95%.
These agrees with the shaded color area in the asymm plot. However, as the name implies,
it should be the 1st variable print 95%, and 2nd 68%. Now it's
opposite. You can also see from the cpvlimit log actually give
68% when it tries to give 95% one.

> Asymmetry posterior (X-Y, integrating over X and Y):

>  Integral asymmetry posterior area 0.95 from -0.5756 to 0.0714