[Hackathon] feat: add AI Workflow Copilot wizard for guided workflow generation

[kz930]

Summary

Adds an in-workspace AI Workflow Copilot that helps biomedical researchers
generate Texera workflows through a guided 4-step wizard backed by Claude. The
copilot is embedded directly in the Angular workspace (no separate app) and
follows a strict review-before-apply flow: workflows are generated to a
review panel where the user inspects each operator's properties and why
explanation before clicking Apply to land them on the canvas.

What's in this PR

• 4-step wizard (ai-wizard-panel.component):

1. Analysis goal (EDA / Prediction / Cleaning / NLP / Custom)
2. Data source — **Existing Dataset** (your uploaded Texera CSV) or
   **dkNET Dataset** (curated biomedical schemas)
3. Scientific framework (CRISP-DM / SEMMA / KDD / Custom) — editable template
   injected as soft prompt guidance
4. Guardrails configuration (train/test split, no leakage, mandatory eval,
   no synthetic data — toggleable, injected as prompt instructions)

• Schema-aware generation — data-profiler.service parses the chosen CSV
  with PapaParse and injects real column names, dtypes, null rates, unique
  counts, and sample values into the prompt so the LLM doesn't hallucinate
  columns.
• White-box validator + 3-attempt repair loop (workflow-validator.service
  • workflow-generator.service) — validates JSON structure, operator type
    existence against Texera's live operator catalog (no hand-maintained
    list), required-property coverage from each operator's jsonSchema,
    source/target port matching on links, and duplicate-ID detection. On
    validation failure, the errors are fed back into the next LLM call for
    self-repair; up to 3 attempts total.
• Review-before-apply panel — generated workflows go to a review panel
  (not directly to the canvas). Each operator shows a per-operator why
  explanation, current properties, and missing-required highlighting.
  Users can edit values inline (JSON for nested fields) before clicking
  Apply to commit the workflow to the canvas.

Demo Video

Watch the demo on YouTube

Test plan

• yarn build from frontend/ completes without errors
• Open workspace → wizard panel renders; can step through all 4 steps
• Step 2 → Existing Dataset path: dataset picker opens, selecting a CSV
  resolves a valid /<owner>/<dataset>/v<n>/<file>.csv backend path
• Step 2 → dkNET Dataset path: schema preview shows on selection
• Step 3 → editing the framework template persists and is sent in the prompt
• Step 4 → toggling guardrails updates the prompt sent to the LLM
• Generate → review panel appears (workflow does not auto-land on canvas)
• Review panel shows per-operator why explanation and any missing-required
  fields are highlighted
• Edit a missing-required field in the review panel → Apply → workflow lands
  with the edited value
• Apply → operators visible on the canvas