Minor fixes

[jaykim1213]

Give a longer description of what this PR addresses and why it's needed

This PR fixes response handling, UI spacing, validation, study content, unused assets, and test organization across demo, example, library, and test studies.

demo-html-input

Reduced the gap between the bar chart and the response.

Before

After

demo-form-elements

• Added validation when the confidence value exceeds 100.
• Reduced the number of sidebar Likert items to prevent overflow.
• Corrected form labels, prompts, and default-value examples.

After

demo-choropleth-map

Required participants to select exactly three states before continuing and corrected the task instructions.

demo-vega

• Assigned unique response IDs to the second Vega demo.
• Removed an unused data file.
• Corrected the instructions, description, and Vega-Lite capitalization.

Other demo studies

Corrected text, metadata, obsolete parameters, and invalid asset references in:

• demo-condition
• demo-dynamic
• demo-html
• demo-image
• demo-ranking-widget
• demo-react-trrack
• demo-screen-recording
• demo-training
• demo-upset
• demo-video
• demo-yaml

example-VLAT-adaptive

• Improved empty and missing-answer handling in feedback views.
• Added fallbacks for missing questions and trials.
• Improved visualization alt text.
• Corrected the documented VLAT question count and removed an unused consent file.

example-brush-interactions

• Improved cleanup of debounced callbacks and brush event listeners.
• Prevented null brush selections from being processed.
• Corrected task wording and response labels.
• Removed unused help and data files.

example-cleveland

Corrected study text and fixed the stacked bar chart height value.

example-llm-chatbot

• Added a note explaining the proxy and allowed-origin requirements.
• Corrected the Trrack action name used for message updates.
• Improved setup instructions.

example-mvnv

• Added requiredValue: "Accept" so participants can continue only after accepting the consent.
• Corrected task instructions, descriptions, HTML, and documentation paths.
• Updated answer posting to support both array and scalar answers.
• Replaced unsafe task-number parsing and handled task number 0 correctly.

test-randomization

Moved the Next button below the stimulus and removed the duplicated introduction page.

Before

After

test-step-logic

Removed unused components, assets, and an unrelated correctAnswer.

Other test studies

Corrected titles, descriptions, labels, formatting, and obsolete references in:

• test-audio
• test-device-restriction
• test-library
• test-likert-matrix
• test-parser-errors
• test-skip-logic

The intentionally invalid parser fixtures remain unchanged in behavior.

library-previs and library-beauvis

Added a small margin between the charts and questions.

Before

After

Documentation and tests

• Corrected repository and study documentation links and wording.
• Moved existing unit tests into sibling tests/ directories.
• Updated the affected E2E tests.

[github-actions]

A preview of 140d3d5 is uploaded and can be seen here:

✨ https://revisit.dev/study/PR1349 ✨

Changes may take a few minutes to propagate.

[JackWilb]

Reduced 7 response options to 6 to prevent overflowing

Likert scales usually have odd number of options so 5 is better

allowing zero dots caused the response UI to break

Probably this is an issue with truthyness. We should fix this properly instead of forcing 1.