ADD: Addition of the Profile Fit PBL height estimate

[jrobrien91]

This PR implements the Profile Method PBLH scheme, where a backscatter profile is fit to an idealized profile via an error function using non-linear least-squares optimization.

References:

Steyn, D. G., M. Baldi, and R. M. Hoff, 1999: The Detection of Mixed Layer
Depth and Entrainment Zone Thickness from Lidar Backscatter Profiles.
J. Atmos. Oceanic Technol., 16, 953–959,
https://doi.org/10.1175/1520-0426(1999)016<0953:TDOMLD>2.0.CO;2.

Sawyer, V., and Z. Li, 2013: Detection, variations and intercomparison of
the planetary boundary layer depth from radiosonde, lidar and infrared
spectrometer. *Atmos. Environ.*, **79**, 518–528,
https://doi.org/10.1016/j.atmosenv.2013.07.019.

Why

While the gradient, tucker and wavelet schemes each have their own methodology for PBLH height, the Profile Method is designed to capture the growth and decay of the mixed-layer. Through a combination of the Profile height and other methods within ACT, the full lifecycle of the mixed layer can be studied.

Validation

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

• [X ] I ran pre-commit or ruff

• [X ] 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 (and claude code) was used iteratively for developing a plan and improvements to retrieval.
  • GPT-5.6 Sol was used for brainstorming and literature review assistance.
  • VS Code - Copilot used for filling in comments that were approved by the user
• Tasks delegated to AI:
  • GPT-5.6 Sol was used to generate a plan to implement Steyn, Baldi and Hoff 1999.
  • Within Juypter Notebooks, Claude Opus 5 was used interactively to improve the functionality of the GPT-5.6 plan. Claude Opus 5 was used to create HTML files to visualize the background material for fitting profiles via the error function. User and Claude Opus 5 interactively tested/improved the methodology while comparing against five months of ARM observations.
  • Claude Opus 5 wrote the tests for the user function, which were verified by the user
• Manual verification performed:
  • User incorporated code from Jupyter Notebooks into pbl_lidar.py, often interacting with both GPT-5.6 SOL and Claude Opus 5 to explain methodology and reasoning. Claude Opus 5 tends to write python to consolidated for my liking, so in many instances, user expanded to be more verbose/readability.

[zssherman]

Resolved the new conflicts from the merge, rerunning the tests