bugs to look at before release (at least these)

[doutriaux1]

• [x] #116 Seg fault on ubuntu [confirmed]
• [ ] #117 black pngs [possible solution needs confirmation]
• [x] #147 mac screen resolution [awaiting vtk fix]
• [ ] #129 can't generate png greater than screen if bg=0 [decided on issuing a warning]
• [ ] #39 Plot does not look right when using vnc or x-forwarding [???]
• [x] #13 put_png_on_canvas disappears if call to canvas.update() [check with @danlipsa if his otherr png related patch might have fixed this]
• [ ] #115 template drawLinesAndMarkersLegend misleading doc [double check but might be fixed]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2135 extra objects created not cleaned? [???]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2114 vcs.getxyvsy() doesn't get the default object [needed fo release?]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2109 vcs.Canvas.isinfile return value consistency [???]
• [x] https://github.com/UV-CDAT/uvcdat/issues/2082 text alignement
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2073 E=mc2 . [needs jupyter notebook but works]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2059 lambert azimuthal proj seem to ignore user custom parameters
• [ ] https://github.com/UV-CDAT/uvcdat/issues/2040 creating pngs in background mode leaks (much less) memory [actually hurts our devel here, but might also be 2135 and 1424]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/1982 Markers are poorly scaled
• [ ] https://github.com/UV-CDAT/uvcdat/issues/1975 vector plot do not have a legend [@aashish24 update on this?]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/1833 #64 Plotting CCSM4 areacello data fails
• [ ] https://github.com/UV-CDAT/uvcdat/issues/1867 DV3D/VCS plot error [i don't think stops release]
• [ ] #73 iso plot failure [???]
• [ ] #74 x.plot(variable) fails with poorly defined bounds [cdms issue?]
• [ ] https://github.com/UV-CDAT/uvcdat/issues/1424 vcs.plot(iso) creates text object that are not removed after clear
• [ ] UV-CDAT/uvcdat#1295 #86 vcs cannot go back into interact mode
• [ ] cdms esmpy regridder pcmdi metrics has new results we need to underrstand why