feat: generalize persona workflow task schema

[joy7758]

Summary:

• add a generic task input schema for the persona workflow demo
• normalize and validate task inputs before running the workflow
• support multiple task types with sample inputs for market research, product design, and UX review
• keep backward compatibility with the earlier product/features demo format

Validation:

• validated market_research_task.json, product_design_task.json, and ux_review_task.json against the schema
• ran python3 demos/persona_workflow_demo.py --task-input demos/product_design_task.json
• ran python3 demos/persona_workflow_demo.py --task-input demos/ux_review_task.json
• ran python3 demos/persona_workflow_demo.py --task-input /tmp/pop_legacy_task.json

Notes:

• this PR is stacked on top of feat: integrate dynamic task input into persona workflow #12 because the earlier workflow branches are not merged into main yet

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a7662edeb2

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Normalize task_type before storing normalized input

validate_task_input trims whitespace when checking membership, so a value like "market_research " passes validation, but normalize_task_input stores the untrimmed value here and downstream code uses it directly (profile_for(task_type) in build_design_output), which raises KeyError at runtime. This means a task can be marked valid and still crash the workflow; trimming task_type during normalization (in both branches) avoids this mismatch.

Useful? React with 👍 / 👎.