Refactor to add answer grid & assoicate llm parsing

Fix: Clarify default prompt — non-dash answers are wrong answers

Simplify: send raw data to LLM, let user's prompt control interpretation

Fix: Treat '-' and similar markers as correct answers in student data

Fix: Handle missing teacher answers — include all student answers instead

Add final prompt preview tab with real student data substitution

Redesign Step 2: per-student report generation with wrong-answer filtering

Fix: Add prompt size check and auto-truncate students to prevent empty LLM responses

Redesign ClassLens v2: 3-step exam analysis workflow