XSPEC bug patches 12.8.0e-j

We have made available a number of XSPEC bug patches/enhancements over the last couple of months and these are summarized below. They can be downloaded from the usual site.

12.8.0e	This adds a couple of new features for PyXspec: interfaces for the parallel and statistic test commands which were introduced in standard XSPEC 12.8.0. The parallel get/set options are accessed through PyXspec's Xset.parallel object, ie: Xset.parallel.leven = 4 # Set parallel processes for Lev-Marq fitting. Xset.parallel.error = 2 # Set parallel processes for error command. Xset.parallel.reset() # Restore single-process mode for all contexts The statistic test command option is accessed in PyXspec through the new Fit.statTest attribute: Fit.statTest = "ad" Fit.statTest = "chi"
12.8.0f	An array access error in the plot goodness command can cause a crash in some cases. Our thanks to Stefano Bianchi for pointing this out.
12.8.0g	Improvements made for running the Goodman-Weare algorithm in the chain command. It now writes keywords CHAINTYP and NWALKERS to the output chain file to distinguish it from those run with the Metropolis-Hastings algorithm. It also fixes bugs that prevented both appending to and re-loading pre-existing Goodman-Weare chain files. Our thanks to Giacomo Vianello for bringing this to our attention.
12.8.0h	When the lrt Tcl command script is run with multiple spectra and background files loaded, the first spectrum is wrongly assigned the background file of the last spectrum. Our thanks to Ed Kellogg for pointing this out.
12.8.0i	This updates the AtomDB to version 2.0.2 (used by the apec models). It also fixes a bug in the identify command, which was preventing the changing of the AtomDB version by way of the APECROOT variable setting.
12.8.0j	This adds a new test option to the tclout statistic and tclout statmethod commands: tclout statistic test tclout statmethod test These will output the current test statistic value and test method respectively. If the 2nd tclout argument is missing or is fit, these will output the fit statistic and fit method (as before).