Fix an issue where toolhead can crash on repeat homing because safe_z is being blindly overwritten with current z position

[jmeier5261]

Address an issue with blindly assuming that if Z is already homed then it is above safe_z which was overwriting the value of safe_z to the current Z position when re-homing if Z was already homed, regardless of whether the current position was actually above the safe_z. Additionally included small typo fixes for logs.

Use case here is that because the machine is in a homed state, if my print fails or ends or any other action leaves my toolhead below specified safe_z height, the machine will not be returned to the actual safe_z when re homing because the code added in ac98aef will overwrite the user variable safe_z with whatever the current Z position is. Ideally we would want to do this only if the current Z is actually above configured safe_z

[jmeier5261]

I want to mention that I am unsure if the change which was added that caused this is actually necessary...it seems that at every point after this, safe_z is only considered if current z position is below it so I'm not sure what the point of setting safe_z to the value of current Z. It may be better just to remove the block itself? I didn't test that approach sufficiently to say. I did notice in the recent fix to that statement #244, the comment on the PR doesn't match the contents of the line itself: so not sure if this was intentional at all.

[jlas1]

Thanks for this, let me check it out

[islasa1]

Thank you so much @jmeier5261 !

I thought I was going crazy, searching online for what might be causing it. Looking at the logic it makes no sense, but as this was my first time setting up Klipper / Klicky I wasn't sure if I just had a setting wrong. It seems like the original intent was to avoid having to move the Z axis if it can be safely assumed to be below the safe_z height.

The if statement added here would do the above. Just tested my setup with the changes as well, and no more crashing into the Z endstop! 😬