Hinted signals for the ion chambers

[canismarko]

This PR changes the signals that are read from an ion chamber during scans. To cut down on the clutter, only the signal net_current will be included in hints. The rest of the readable signals will still be read, but not hinted.

It also might fix a race condition in reading the scaler. Since the PVs update as the scaler reads, it's possible that the cached value that the IonChamber device has is stale by the time the scaler triggering is done. This showed up as an occasional low blip in the time signal (scaler channel 0). Since this time signal is used to calculate current, this meant the current is not reliable.

This PR fixes this by using the uncached signal formats introduce in ophyd-async 0.8.0. The downside is that we need to upgrade to ophyd-async 0.8.0 which isn't released yet. Most of the changes in this PR are actually for compatibility with ophyd-async 0.8.0.

Things to do before merging:

• add tests
• flake8, black, and isort

[Cathyhjj]

I like the hint that only includes net-count; maybe we should also reduce hints for monochromator; right now, all the energy related such as offset/vert/ID... are hinted. In practice, maybe only 'energy' or 'energy-monochromator-energy' should be hinted. But this should be done in a different branch & PR I guess

[canismarko]

Thanks, @Cathyhjj

I agree we should revisit the energy positioner's hints. Ophyd-async 0.8.0 changed how those signals are determined, so we can see what actually happens once we merge this PR.