The SummaryRecord.add_value() method is extended to record the data of
MindExplain.

change ImagefolderDV2 name

change ds.transforms.vision to ds.vision

change batch api to match map api more closely

compose op changes

test_pylint

remove compose op from vision, move to transform module, refactor map and batch to use column_order

Merge pull request !5389 from ougongchang/fix_summarycollector

- handle collection for multiple trains
- how many tensors to collect when sunk
- change loglevel for get_learning_rate
- update calculation of `max_file_size`
- fix how collect_tensor_freq counting

- fix writer pool not exit when max_file_size too small
- fix API docs to illustrate `collect_tensor_freq` and `max_file_size`

We collect graph and dataset graph in begin stage before,
If there compile graph fail in GPU, we also collect graph
and dataset graph to summary dir, it will confuse user.

So we collect graph and dataset graph in step end stage now,
If there compile graph fail, we will not collect graph and dataset
graph.

When `collect_tensor_freq` is specified as `None`,
the `collect_tensor_freq` would be auto calculated.

The previous behavior is to collect at most 50 steps,
now changing to 20

1. collect the origin network in model, and set it to cb_params
2. collect the origin network name in SummaryCollector
3. Update the SummaryCollector API Doc

The useage is error when collect custom lineage data.

Add more detail note for SummaryCollector and SummaryRecord,
else if it is used not right, some proplem will be caused.

fix get loss error when it not a scalar and fix process specified data
failed when the action is False, and collect_specified_data parameter is
not None

Merge pull request !2743 from ougongchang/summary_collector_ut

Has fixed collecting optimizer error when mode is eval

Before, we only decide whether to collect data by current step,
it will not work well in dataset sink mode, so we check to see
if it's a dataset sink mode, and decide whether to collect data.

I added a SummaryCollector to help users automatically collect information
such as the network, loss, learning rate and so on, making it easier to collect this information.

It also can collect train lineage and eval lineage information which is
collected by TrainLineage Callback and EvalLineage Callback in
MindInsight.

I also add some UT for SummaryCollect to keep the code correct.