Debugging option to understand serialization issues

Nobody has claimed this yet.

### Use Case

This is a result of: https://github.com/influxdata/telegraf/issues/17235#issuecomment-3509885081

When serialization fails, it would be handy to have an option to view the ingested data to better understand where the serialization issue is coming from in detail. This may be required for cases of the `inputs.exec` plugin or alike, where data is potentially dynamically generated.

In our specific case, we use a PowerShell script that outputs information from Veeam as a JSON object, which is ingested in Telegraf and written in Influx Line Protocol to our server. But because Powershell can be quirky, sometimes it may run into an issue where the object ends up potentialyl malformed. Being able to see _what exactly_ was malformed may help us to improve that script. There are likely other use-cases also.

### Expected behavior

The expected behaviour would be that given an option (either globally in the `agent` settings or more tightly scoped at the plugin configuration level) the agent would log serialization issues and possibly write them to it's own default logging location (`stdout` in most cases, much like most other errors).

### Actual behavior

None, it's a feature request. :)

Right now, this behaviour does not exist.

### Additional info

_No response_

Start by tracing the serialization failure path for inputs.exec and reviewing how agent settings, plugin configuration, and stdout logging are handled. Define an option that exposes the ingested data when serialization fails, with behavior scoped globally or to a plugin, and confirm that the malformed input is logged for diagnosis.

Written by the indexing model from the issue text.