fix: log if an error is returned from upstream

[btasker]

Summary

Raising to try and address #16386

Because of the partial writes handling, the influxdb_v2 output plugin doesn't complain if it receives a 400. This can mask configuration issues and make it look like data is being written when it isn't.

This PR adjusts the handling to ensure that a 400 matches the partial writes behaviour (i.e. includes a message which begins with invalid) and generates a logline if that's not the case.

Note: it doesn't change behaviour beyond that - we likely still want the metrics to be discarded so they don't block the buffer. But, this also means that, as written, Telegraf will print both a failure and a success message:

2025-01-09T12:18:27Z W! [outputs.influxdb_v2] failed to write metric to testingdb (will be dropped: 400 Bad Request)
2025-01-09T12:18:27Z D! [outputs.influxdb_v2] Wrote batch of 13 metrics in 3.996059ms

Checklist

• No AI generated code was used in this PR

Related issues

resolves #

[Hipska]

This seems already be fixed by #16388?

[btasker]

Yep, Sven and I were talking in the background and I forgot to come back and close this out