Improve "error_text" message in query prepare

As discussed in https://gitlab.cern.ch/cta/operations/-/issues/1008, the message passed through the "error_text" field of the xrdfs query prepare command can be very opaque...

Therefore, we need to work on improving it.

The following steps need to be taken:

• Check existing xattrs for a tape file.
• Add any new xattrs that can help enriching the "error_text" message in the query prepare command.
  • Possible new attributes:
    • Error timestamp
    • Error type
    • Error origin (cta-taped, cta-frontend, EOS, etc.)
    • Etc...
  • Example of enriched "error_text" message:
    • "[123456789.00][cta-taped] trigger workflow - synchronous workflow failed"
• Review default "error_text" messages filed by EOS. Check if they are correctly adapted to each specific failure scenario.
• Review mechanism that passes messages from CTA to EOS (&mgm.errmsg= field).