feat: M2 AI tips — LiteLLM gateway, context assembler, end-to-end generation pipeline

Issues closed: #86, #87, #88, #89, #90, #91, #79, #80, #82

infra:
- docker-compose `ai` profile: Ollama + LiteLLM services
- infra/litellm/litellm_config.yaml: tip-generator / embedder / judge aliases
- .env.example: LITELLM_URL, LITELLM_MASTER_KEY, OLLAMA_URL

ml/serving:
- POST /generate: calls LiteLLM tip-generator alias, returns TipCandidate[]
- JSON retry loop (2 retries with correction prompt on malformed response)
- _parse_llm_json strips markdown fences

ml/features:
- context.py: build_context() assembles user signals → PromptContext
  (sorts overdue/high-priority tasks first for LLM prompt quality)

shared-types:
- TipKind, TipSource, TipCandidate types
- Tip gains kind + rationale fields

services/api:
- recommender: 3-stage pipeline (assemble → score → serve)
  Stage 1: Todoist tasks + LLM candidates fetched in parallel
  Stage 2: egreedy bandit scores merged candidate pool
  Stage 3: serve + log with prompt_version, llm_model, tip_kind
- tip_scores: prompt_version, llm_model, tip_kind columns + migrations
- config: LITELLM_URL added
- integrations: surface token_status in /integrations response

tests:
- ml/serving/tests/test_generate.py: 13 tests (retry, 502/503, fence variants)
- ml/features/test_context.py: 9 tests (sorting, edge cases)
- services/api recommender.unit.test.ts: 16 pure-function tests (inferReward, dueAgeDays)
- services/api recommender.test.ts: 4 integration tests (tip_scores columns, LLM fallback)
- shared-types: TipCandidate, rationale, full TipFeedback action set

docs:
- ADR-0008: LiteLLM AI gateway decision
- overview.md: M2 pipeline description updated
- ml/README.md: serving + features roles updated

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

ml/features/context.py

@@ -0,0 +1,63 @@
+"""
+Context assembler — converts raw user signals into a PromptContext for LLM tip generation.
+
+Usage:
+    from ml.features.context import build_context
+    ctx = build_context(tasks, hour_of_day=9, day_of_week=2)
+"""
+
+from __future__ import annotations
+from dataclasses import dataclass, field
+
+
+@dataclass
+class TaskSignal:
+    id: str
+    content: str
+    priority: int = 1           # 1–4 (Todoist scale)
+    is_overdue: bool = False
+    task_age_days: float = 0.0
+    due_date: str | None = None
+
+
+@dataclass
+class PromptContext:
+    tasks: list[dict] = field(default_factory=list)
+    hour_of_day: int = 12
+    day_of_week: int = 0
+    extra: dict = field(default_factory=dict)
+
+
+def build_context(
+    tasks: list[TaskSignal],
+    hour_of_day: int = 12,
+    day_of_week: int = 0,
+    extra: dict | None = None,
+) -> PromptContext:
+    """
+    Assemble user signals into a PromptContext.
+
+    Signals are sorted so overdue + high-priority tasks appear first,
+    giving the LLM the most actionable context at the top of the prompt.
+    """
+    sorted_tasks = sorted(
+        tasks,
+        key=lambda t: (not t.is_overdue, -t.priority, -t.task_age_days),
+    )
+    task_dicts = [
+        {
+            "id": t.id,
+            "content": t.content,
+            "priority": t.priority,
+            "is_overdue": t.is_overdue,
+            "task_age_days": round(t.task_age_days, 1),
+            "due_date": t.due_date,
+        }
+        for t in sorted_tasks
+    ]
+    return PromptContext(
+        tasks=task_dicts,
+        hour_of_day=hour_of_day,
+        day_of_week=day_of_week,
+        extra=extra or {},
+    )