v1.1.0

Added

• include_extra_columns (per config entry, off by default). The response set was fixed, so a
  view carrying columns of its own — a label, a unit, a formatted timestamp — had them dropped,
  because the reader never selected them. With the option on, every remaining column of the queried
  table comes along and can feed a table view directly.

  Column names come from information_schema through _fetch_table_columns, so they are by
  definition real columns of that table, and they are quoted as well. Reserved names (entity_id,
  value, state, time, bucket, minute) are never selected twice. With downsample > 0 each
  extra column is wrapped in last() so GROUP BY stays valid.

  With the option off the generated SQL is character for character what it was, so existing
  dashboards cannot be affected.

• sum as a downsample_method, alongside avg and last. For rows that are already a
  quantity per bucket — a cost, or a number of kWh per hour — asking for daily buckets should add
  those hours up rather than average them or take the last one. Only used when the caller asks for
  it explicitly; the automatic choice is unchanged.

Documentation

• README.nl.md added, with language badges on both READMEs, matching the other integrations.
• New sections on extra columns and a worked example: daily peaks presented as a readable table.