fix "stale pointer" bug when re-opening a subpatch on a scalar that has an event bound to it. The bug was that the scalar symbol wasn't getting unbound because scalar_vis wasn't getting called.

try removing the artifacts to see if it caused a regression with the Ubuntu runners hanging on exit (I can't debug it through gitlab because of its arbitrary limit on build log size... :(

fix tar_em_up.sh to store binaries in the main repo directory so that the gitlab ci can upload them as artifacts, and change the ci config to find them

Get rid of some really annoying legacy Tcl commands in the core. As long as we don't have any LED indicators in the console window, show an error for audio dropouts instead. Also get rid of some debugging messages in the pddp.

add feature to "normalize" svg path data and pass it along as a FUDI message (e.g., "M0-2l3 4" becomes "M 0 -2 l 3 4")
This gets triggered when the user include the "d=" and double quotes as the object args to [draw path].  So [draw path d="M0-2l3 4"] triggers a conversion to [draw path M 0 -2 l 3 4], whereas [draw path M0-2l3 4] does not. This is necessary because Pd lets the user include scalar fields inside path data, and we can't normalize that.

Add a check that a toc entry actually exists before adding it. Added DISIS and Gem entries back to the toc.