BUG - Hi L2 - problems propegating error and uncertainty in maps

[tmplummer]

Description of the issue

I found two problems with latest test Hi maps:

• The change to using linear interpolation caused NaNs in statistical uncertainties to be set to zero.
• Systematic errors for pixels with zero exposure time are zero when they should be NaN.

Steps to reproduce the issue

No response

Expected vs Actual behavior

No response

Code Snippet (If applicable)

Code

Additional notes, affected areas, and suggested fixes

No response