Resume dim for non i2cs devices

[lnr0626]

Proposed change

Enables the resume_dim flag for non i2cs devices

Additional information

• This PR fixes or closes issue: fixes Enable Resume Dim Feature on non-I2CS Devices #415

Checklist

• The code change is tested and works locally.
• Local tests pass. Your PR cannot be merged unless tests pass
• Tests have been added to verify that the new code works.
• Code documentation was added where necessary

I only have i2cs devices, so will need help verifying this works properly for non i2cs devices :-)

[krkeegan]

Awesome, I appreciate you making that a separate method so that it can be reused for other flags.

[krkeegan]

My only hesitation here is the addition of the get-engine request to the refresh request. I see it is limited to only occur in cases where the engine is unknown or refresh is called with force.

I understand the benefit of this, making things "just work" for users. The concern that I have always had, is as we chain more commands to refresh we run a greater risk of complications.

I think this is probably a good addition, but I want to think about it for a minute.

[lnr0626]

that makes sense - I added it mainly because i had a few devices that didn't have that set initially, and without that the devices likely wouldn't respond correctly.

An alternative would be to add some additional handling so that if we don't know the engine and we received a NACK (invalid checksum) in response to a set_operating_flags message, we either update the engine to i2cs and retry; or run get_engine and then retry

my main hesitation with just setting the engine to i2cs is that's a fairly large assumption that may break at some point.

that might also have the potential for infinite loops if (for some reason), we are never able to set the engine?

[krkeegan]

I tested out refresh with the get_engine command on a few i2cs devices and some old i1 devices and it all works fine. We already do a model request, so it likely makes sense to get the engine version too.

A few small additions, but I am otherwise happy to merge this.

[lnr0626]

those changed looked good to me, so just went ahead and commited them. thanks for the reviews and the help with testing this stuff out!