PERF: delay preset loading for hutch-python

[tangkong]

Description

• Utilizes PERF: delay preset sync until tab completion / attribute access, track file desync pcdsdevices#1317 to defer preset loading

Motivation and Context

We want to defer preset loading for performance reasons.

This will fail tests until the corresponding pcdsdevices PR is merged.

How Has This Been Tested?

I was using this to test presets while developing pcdshub/pcdsdevices#1317

Where Has This Been Documented?

This PR

Left: Before, Right: After

In getting that screenshot I ran into a variety of load times, preset deferral was always better but by varying margins.

Ran on a clone of XCS's hutch-python profile, copied into my dev area

Pre-merge checklist

• Code works interactively (a real hutch config file can be loaded)
• Code contains descriptive docstrings, including context and API
• New/changed functions and methods are covered in the test suite where possible
• Test suite passes locally
• Test suite passes on GitHub Actions
• Ran docs/pre-release-notes.sh and created a pre-release documentation page
• Pre-release docs include context, functional descriptions, and contributors as appropriate

[tangkong]

I guess there are no real tests to do here, since this is a pcdsdevices feature.

[ZLLentz]

Does this implicitly change our minimum supported pcdsdevices version? Is this something we would address in the conda forge feedstock?

[tangkong]

Ah you're right. That slipped my mind.

This will technically rely on a pcdsdevices version that doesn't exist yet. I'm reluctant to tag pcdsdevices right now because we might accumulate more changes, but once I change the pins the tests will fail. I'll move this back to WIP until we're ready to more forward.

More incentive for me to look at pyca / typhos I guess

[ZLLentz]

I found this again in my PR reviews backlog. Maybe we should tag pcdsdevices next week.

[tangkong]

Ah yes. This also reminds me of the whole tagging/new-conda-effort we had going a month ago

[tangkong]

This will fail until we've tagged/deployed pcdsdevices, I've updated the pin here slightly prematurely

[tangkong]

Apparently in py3.12 flake8 started linting strings, a problem that was fixed with 6.1.0. I suspect that shifting to ubuntu latest shifted the default python version to 3.12, exposing us to this error.

I verified that flake8 v6.0.0 and py3.12 reproduces this

Updating the flake8 version in the pre-commit config fixes this

[tangkong]

This is good to go now, as long as we're sure that we're ok with ipython being unpinned

[ZLLentz]

I'm happy with this