feat(ml): multi-agent context framework + v4 orchestrator prompt

Adds ml/agents/ — five specialised sub-agents (overdue_task, momentum,
time_of_day, recent_patterns, focus_area) each producing a prompt snippet
from user signals. A registry wires them up; the orchestrator prompt in
ml/serving/prompts.py synthesises their outputs into one tip via LiteLLM.

Also wires /api/agents route in the API and updates the Dockerfile to copy
the full ml/ tree with PYTHONPATH=/app so agent imports resolve correctly.

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

ml/agents/momentum.py

@@ -0,0 +1,49 @@
+from __future__ import annotations
+from typing import ClassVar
+from .base import BaseAgent, AgentInput, AgentOutput
+
+
+class MomentumAgent(BaseAgent):
+    """Characterises the user's recent engagement trend from profile features."""
+    agent_id: ClassVar[str] = "momentum"
+    ttl_seconds: ClassVar[int] = 21600  # 6h
+    version: ClassVar[str] = "1.0.0"
+
+    def compute(self, inp: AgentInput) -> AgentOutput:
+        completion = inp.profile.get("completion_rate_30d")
+        dismiss = inp.profile.get("dismiss_rate_30d")
+        volume = inp.profile.get("tip_volume_30d")
+
+        parts: list[str] = []
+
+        if completion is not None:
+            pct = round(completion * 100)
+            if pct >= 50:
+                parts.append(f"The user completes {pct}% of tips (strong engagement).")
+            elif pct >= 25:
+                parts.append(f"The user completes {pct}% of tips (moderate engagement).")
+            else:
+                parts.append(
+                    f"The user completes {pct}% of tips "
+                    f"(low engagement — prefer simple, immediately actionable tips)."
+                )
+        else:
+            parts.append("No completion-rate data yet (new user).")
+
+        if dismiss is not None:
+            dpct = round(dismiss * 100)
+            if dpct >= 40:
+                parts.append(f"Dismiss rate is high ({dpct}%) — avoid repetitive or irrelevant tips.")
+            elif dpct <= 10:
+                parts.append(f"Dismiss rate is low ({dpct}%).")
+
+        if volume is not None and int(volume) < 5:
+            parts.append("Very few tips served so far — this is an early-stage user.")
+
+        prompt = " ".join(parts) if parts else "No engagement data available yet."
+        snapshot = {
+            "completion_rate_30d": completion,
+            "dismiss_rate_30d": dismiss,
+            "tip_volume_30d": volume,
+        }
+        return self._make_output(inp, prompt, snapshot)