Update log statement with a time that prints

[jeanconn]

Description

Fixes log statements like:

2023-08-10 08:31:15,261 get_cmds: Getting commands from <chandra_time.Time.DateTime object at 0x7fa1eb0053f0> to <chandra_time.Time.DateTime object at 0x7fa1eb005600> for scenario=Non
e

Interface impacts

Testing

Unit tests

• [x] Mac
• [x] Linux

Independent check of unit tests by [REVIEWER NAME]

• [ ] [PLATFORM]:

Functional tests

On master at eb5868

In [1]: import kadi.commands.commands_v2
In [2]: kadi.commands.commands_v2.logger.setLevel('INFO')
In [4]: from Chandra.Time import DateTime
In [5]: cmds = kadi.commands.commands_v2.get_cmds(DateTime("2023:001"), DateTime("2023:005"))
2023-10-20 09:46:46,002 get_cmds: Getting commands from <chandra_time.Time.DateTime object at 0x1179b62f0> to <chandra_time.Time.DateTime object at 0x1179b7550> for scenario=None
2023-10-20 09:46:46,007 get_cmds: Getting commands from archive only because of: scenario=None before_recent_cmds=True HAS_INTERNET=True

On this PR at 507ad83

In [1]: import kadi.commands.commands_v2
In [4]: kadi.commands.commands_v2.logger.setLevel('INFO')
In [9]: from Chandra.Time import DateTime
In [10]: cmds = kadi.commands.commands_v2.get_cmds(DateTime("2023:001"), DateTime("2023:005"))
2023-10-20 09:46:19,730 get_cmds: Getting commands from 2023:001:00:00:00.000 to 2023:005:00:00:00.000 for scenario=None
2023-10-20 09:46:19,732 get_cmds: Getting commands from archive only because of: scenario=None before_recent_cmds=True HAS_INTERNET=True

[taldcroft]

I'd suggest fixing the upstream code.

[jeanconn]

I've been using WIPs to keep track of things I'd like to move forward more than issues. This is totally untested and wasn't for review.

[taldcroft]

I thought DateTime had a decent str repr now, but apparently not.

[jeanconn]

Also, what you consider the "upstream code" for this?

[taldcroft]

Also, what you consider the "upstream code" for this?

Whatever is calling this function. But I just suggested another option.

[jeanconn]

"Whatever is calling this function." Sure -- though from my perspective that would just be kadi_validate_states which seemed a big tree to traverse. So yes, I agree that isolating to just have a generic solution that prints something reasonable at this line in the code makes sense to me.

[taldcroft]

This just needs some functional testing.

[taldcroft]

And flake8, apparently.