The code are wrapped due to us using line-length=79 before, but now some
of the code can easily fit in 88 chars.
The formatter did not change this automatically when we changed line-length,
because of magic trailing commas that gets added when they turn into multiple lines.
Black/ruff wrapping them in parens and splitting them into multiple lines
was making it difficult to read.
* standardise plot data output
* add data required for new anchors to renderer properties
* feed dvc-render updates into API output
* update --json --split api with new data types (send split anchor definitions as strings)
* accomodate zoom and pan anchor
* switch anchor_revs to revs_with_datapoints
* accomodate height and width anchors
* drop terrible idea of holding all data as strings
* fix plot having multiple x fields
* move inferred properties off vega converter class
* plots: shorten field names
* plots: handle missing field names
* plots: make compatible with py3.8
* plots: fix for missing data
* plots: add test for field shortening
* plots: strip separators regardless of os
* plots: don't shorten field name
* plots: drop test for shortening field name
* plots: vega: converter: support multi series data split and props inference
Fixes: #7754, #8399
Related: iterative/studio#4369
* plots: converter: make converter infer x and y labels if not provided
* plots: vega: converter: roll back to datapoints format
* plots: vega: converter: refactor
* plots: infer labels on
* plots: get_plot_id: fix naming
* plots: vega: converter: notes regarding converting
skshetry
Matt Seddon
Matt Seddon
Dave Berenbaum
David de la Iglesia Castro
Paweł Redzyński