Add multi-selection support to dropdown widget

[wdanilo]

The WidgetMultiSelection should work almost the same way as WidgetSelection, but (obviously) allow selecting multiple choices (it does not close after click at choice), and the picked items are put in List - and this list should be displayed with vector widget, because we want to allow reordering picked choices.

---

This task is automatically imported from the old Task Issue Board and it was originally created by Paweł Grabarz.
Original issue is here.

Step 3: Support for multi-selection

Delivery requirements:

The implementation should be refined to allow a multi-selection of entries. The information if the widget should allow multi-selection should be delivered from the engine together with the information that a particular input should be displayed as a drop-down widget.

See the widget metadata design to learn more: https://github.com/enso-org/design/blob/main/epics/widgets/drop-downs/dynamic-widgets.md

[farmaazon]

The one difficulty I can think of is passing configuration of picked options to vector elements (each element could have different option picked, and thus different config). However, as I remember, it is the WidgetFunction now (not WidgetSelection), which picks of which option a configuration should be applied.

[farmaazon]

Refinement notes:

• Recognizing what options are selected should be simple text matching of vector's elements code (like it is in single-choice right now).
• Now we assume the widget always produces a vector (it may try to read non-vector as a single value, but any change would turn it to vector)
• Let's also add API for Vector AST node and use it in WidgetVector as well (see TODO there)

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-03-18):

Progress: Implemented multiselect support in selection widget. It should be finished by 2024-03-22.

Next Day: Next day I will be working on the #5138 task. Add multi-choice mode to vector widget; e2e testing.

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-03-19):

Progress: Still working on testing. It should be finished by 2024-03-22.

Next Day: Next day I will be working on the #5138 task. Finish e2e tests.

[enso-bot]

Keziah Wesley reports a new STANDUP for yesterday (2024-03-19):

Progress: Addressed backend/most testing issues. Working on maintaining dropdown state when a selection is made and the placeholder widget is replaced with an argument widget. It should be finished by 2024-03-22.

Next Day: Next day I will be working on the #5138 task. Probably finish feature.

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-03-21):

Progress: Working on resumable widgets, split off a PR with the main dropdown implementation. It should be finished by 2024-03-22.

Next Day: Next day I will be working on the #5138 task. Finish up.

[enso-bot]

Keziah Wesley reports a new STANDUP for the last Friday (2024-03-22):

Progress: Refactoring interactions to support resumable widget edits It should be finished by 2024-03-22.

Next Day: Next day I will be working on the #5138 task. Same task.

[enso-bot]

Keziah Wesley reports a new 🔴 DELAY for yesterday (2024-03-25):

Summary: There is 7 days delay in implementation of the Add multi-selection support to dropdown widget (#5138) task.
It will cause 0 days delay for the delivery of this weekly plan.

Delay Cause: implementing

[enso-bot]

Keziah Wesley reports a new STANDUP for yesterday (2024-03-25):

Progress: New interaction model passing tests. Cleaning up some implementation details. It should be finished by 2024-03-29.

Next Day: Next day I will be working on the #5138 task. PR.

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-03-26):

Progress: Merge conflicts. Refactoring to introduce new implementation with less API change for now. It should be finished by 2024-03-29.

Next Day: Next day I will be working on the #5138 task. PR.

[enso-bot]

Keziah Wesley reports a new STANDUP for yesterday (2024-03-27):

Progress: Debugging unit tests that used low-level API. Implemented logic for recognizing a port that is a successor of another (placeholder) port. It should be finished by 2024-03-29.

Next Day: Next day I will be working on the #5138 task. PR.

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-03-28):

Progress: Finished except for one issue in e2e testing. It should be finished by 2024-03-29.

Next Day: Next day I will be working on the #5138 task. Fix time-traveling mocks and start next task.

[enso-bot]

Keziah Wesley reports a new STANDUP for the last Friday (2024-03-29):

Progress: Refactoring to support reactive activity states to prevent bugs caused by explicit tracking of start/stop state getting out of sync with actual interaction It should be finished by 2024-03-29.

Next Day: Next day I will be working on the #5138 task. Fix test issues and start next task.

[enso-bot]

Keziah Wesley reports a new 🔴 DELAY for yesterday (2024-04-01):

Summary: There is 5 days delay in implementation of the Add multi-selection support to dropdown widget (#5138) task.
It will cause 0 days delay for the delivery of this weekly plan.

Delay Cause: Issue involves a lot of design consideration for APIs for collaboration between widgets

[enso-bot]

Keziah Wesley reports a new STANDUP for yesterday (2024-04-01):

Progress: Debugging reactivity. It should be finished by 2024-04-03.

Next Day: Next day I will be working on the #5138 task. Fix test issues and start next task.

[enso-bot]

Keziah Wesley reports a new STANDUP for today (2024-04-02):

Progress: Finished reactive refactor. Documented some new APIs. Debugging unexpected order of events in mocks. It should be finished by 2024-04-03.

Next Day: Next day I will be working on the #5138 task. Fix test issues and start next task.