Fix #110: Ensure spectrogram plotting correctly handles non-UTC time scales

[Amityush-lgtm]

PR Description

Fixes an issue where plotting data with different time scales i.e. 'tt' and 'utc' would result in time offsets because GenericSpectrogram plotting was using the raw time scale components directly via .datetime. Corrects the plotting logic to always convert to UTC first.
Closes #110

What changed

• Modified radiospectra/mixins.py to ensure self.times is converted to UTC before generating datetime objects for plotting.
• Added internal helper _times_utc_datetime to handle the conversion if supported by the time object.
• Validated that plotting correctly aligns with UTC time regardless of input time scale.

@samaloney, Let me know your thoughts on this!!

[samaloney]

This is not specific to this PR. I'm posting this to all open PRs. Please be aware of the update to our AI usage policy, specifically on the disclosure of and acceptable uses of AI.

[samaloney]

Thanks for the RP but I'm not sure defaulting to UTC is the best option. What if someone wants to plot in tt scale, it would be better to convert to the existing scale if one exists.

Should be able to use or follow similar approach to astroy's time_support

[Amityush-lgtm]

Thanks a lot for your response. @samaloney I am currently working on a way to solve the issue by not defaulting to UTC but i am stuck and having a doubt regarding the scales. For mixed time scales on the same axes, should the first plotted scale define the axis scale for the plots?

[samaloney]

Great I think this is almost ready just a small tidy up of the imports and should be good to go.