Allow use of pre-computed lengths when grouping by length. (#10953)
A new argument `length_column_name` has been added to
`TrainingArguments`, with default value `"length"`. If this column
exists and `group_by_length` is `True`, the train sampler will use
it for grouping rather than computing it before training starts.
This is an optimization that allows the user to prepare data for fast
processing, preventing sequential access to the dataset as described in
issue #10909.