ICARTT v2

[jrobrien91]

What changed

Removes the icartt library dependency from ACT, creating an new Icartt class within act/io/icartt.py that allows for reading and writing of FFI 1001 data.

write_icartt is an entirely new function designed to close the loop between ARM netcdf and ICARTT formatted files.

Why

Closes ACT 2.0 roadmap remaining item (write_icartt).
Also, greatly improves testing of the icartt class, particularly the handling of special comment lines

Issues closed

Closes #1051

Validation

Mark each item as done or N/A. Add one short evidence line for each item.

• I ran pre-commit or ruff

• I ran tests and/or added tests

AI usage & manual verification

If AI tools were used, complete this section.

• AI used: Yes
• Model/tooling:
  • Claude Opus 5 via Claude Code
• Tasks delegated to AI:
  • Given the current implementation of the ICARTT reader, the NASA ICARTT v2.0 standards documentation, and a previous generated ICARTT class created by the developer, remove the icartt.py dependency and improve testing suite.
  • Developer removed icartt from the continuous integration and docs environments.
• Manual verification performed:
  • From the ARM AAF ICARTT formatted file within the test data, preformed a loop of reading in the data and re-writing to a new file to test performance. Visual inspection of the before and after ICARTT files. Checked if methodology made sense and generated class was similar to previously created version.
  • joe.ict was included within this markdown for visual inspection against the ARM Test Data repo. This file was generated via write_icartt after reading in the ARM AAF ICARTT file.

Impact note (optional):
- there is not a change for users using the read_icartt function. However, the icartt library dependency is removed.

[jrobrien91]

Hmm not letting me include the written sample .ict file. I'll send over internally

[jrobrien91]

Still need to add the scaling factor function but will do so in the morning; other suggested corrections applied

[jrobrien91]

Scale factor function created for the incredibly rare chance someone wants to use a non-one scale factor. The caveat on this is it does not touch the upper and lower detection limit flags, which will always be -7777 and -8888