feat(email_gen): draft constituent letter generator + bill text indicators

- Add DraftLetterPanel: collapsible UI below BriefPanel for bills with a
  brief; lets users select up to 3 cited points, pick stance/tone, and
  generate a plain-text letter via the configured LLM provider
- Stance pre-fills from follow mode (pocket_boost → YES, pocket_veto → NO)
  and clears when the user unfollows; recipient derived from bill chamber
- Add POST /api/bills/{bill_id}/draft-letter endpoint with proper LLM
  provider/model resolution from AppSetting (respects Settings page choice)
- Add generate_text() to LLMProvider ABC and all four providers
- Expose has_document on BillSchema (list endpoint) via a single batch
  query; BillCard shows Brief / Pending / No text indicator per bill

Authored-By: Jack Levy

backend/app/services/llm_service.py

@@ -183,6 +183,10 @@ class LLMProvider(ABC):
     def generate_amendment_brief(self, new_text: str, previous_text: str, bill_metadata: dict) -> ReverseBrief:
         pass
 
+    @abstractmethod
+    def generate_text(self, prompt: str) -> str:
+        pass
+
 
 class OpenAIProvider(LLMProvider):
     def __init__(self, model: str | None = None):
@@ -218,6 +222,14 @@ class OpenAIProvider(LLMProvider):
         raw = response.choices[0].message.content
         return parse_brief_json(raw, "openai", self.model)
 
+    def generate_text(self, prompt: str) -> str:
+        response = self.client.chat.completions.create(
+            model=self.model,
+            messages=[{"role": "user", "content": prompt}],
+            temperature=0.3,
+        )
+        return response.choices[0].message.content or ""
+
 
 class AnthropicProvider(LLMProvider):
     def __init__(self, model: str | None = None):
@@ -247,6 +259,14 @@ class AnthropicProvider(LLMProvider):
         raw = response.content[0].text
         return parse_brief_json(raw, "anthropic", self.model)
 
+    def generate_text(self, prompt: str) -> str:
+        response = self.client.messages.create(
+            model=self.model,
+            max_tokens=1024,
+            messages=[{"role": "user", "content": prompt}],
+        )
+        return response.content[0].text
+
 
 class GeminiProvider(LLMProvider):
     def __init__(self, model: str | None = None):
@@ -272,6 +292,14 @@ class GeminiProvider(LLMProvider):
         response = self._make_model(AMENDMENT_SYSTEM_PROMPT).generate_content(prompt)
         return parse_brief_json(response.text, "gemini", self.model_name)
 
+    def generate_text(self, prompt: str) -> str:
+        model = self._genai.GenerativeModel(
+            model_name=self.model_name,
+            generation_config={"temperature": 0.3},
+        )
+        response = model.generate_content(prompt)
+        return response.text
+
 
 class OllamaProvider(LLMProvider):
     def __init__(self, model: str | None = None):
@@ -326,6 +354,16 @@ class OllamaProvider(LLMProvider):
             )
             return parse_brief_json(raw2, "ollama", self.model)
 
+    def generate_text(self, prompt: str) -> str:
+        import requests as req
+        response = req.post(
+            f"{self.base_url}/api/generate",
+            json={"model": self.model, "prompt": prompt, "stream": False},
+            timeout=120,
+        )
+        response.raise_for_status()
+        return response.json().get("response", "")
+
 
 def get_llm_provider(provider: str | None = None, model: str | None = None) -> LLMProvider:
     """Factory — returns the configured LLM provider.
@@ -344,3 +382,72 @@ def get_llm_provider(provider: str | None = None, model: str | None = None) -> L
     elif provider == "ollama":
         return OllamaProvider(model=model)
     raise ValueError(f"Unknown LLM_PROVIDER: '{provider}'. Must be one of: openai, anthropic, gemini, ollama")
+
+
+_BILL_TYPE_LABELS: dict[str, str] = {
+    "hr": "H.R.",
+    "s": "S.",
+    "hjres": "H.J.Res.",
+    "sjres": "S.J.Res.",
+    "hconres": "H.Con.Res.",
+    "sconres": "S.Con.Res.",
+    "hres": "H.Res.",
+    "sres": "S.Res.",
+}
+
+_TONE_INSTRUCTIONS: dict[str, str] = {
+    "short": "Keep the letter brief — 6 to 8 sentences total.",
+    "polite": "Use a respectful, formal, and courteous tone throughout the letter.",
+    "firm": "Use a direct, firm tone that makes clear the constituent's strong conviction.",
+}
+
+
+def generate_draft_letter(
+    bill_label: str,
+    bill_title: str,
+    stance: str,
+    recipient: str,
+    tone: str,
+    selected_points: list[str],
+    include_citations: bool,
+    zip_code: str | None,
+    llm_provider: str | None = None,
+    llm_model: str | None = None,
+) -> str:
+    """Generate a plain-text constituent letter draft using the configured LLM provider."""
+    vote_word = "YES" if stance == "yes" else "NO"
+    chamber_word = "House" if recipient == "house" else "Senate"
+    tone_instruction = _TONE_INSTRUCTIONS.get(tone, _TONE_INSTRUCTIONS["polite"])
+
+    points_block = "\n".join(f"- {p}" for p in selected_points)
+
+    citation_instruction = (
+        "You may reference the citation label for each point (e.g. 'as noted in Section 3') if it adds clarity."
+        if include_citations
+        else "Do not include any citation references."
+    )
+
+    location_line = f"The constituent is writing from ZIP code {zip_code}." if zip_code else ""
+
+    prompt = f"""Write a short constituent letter to a {chamber_word} member of Congress.
+
+RULES:
+- {tone_instruction}
+- 6 to 12 sentences total.
+- First sentence must be a clear, direct ask: "Please vote {vote_word} on {bill_label}."
+- The body must reference ONLY the points listed below — do not invent any other claims or facts.
+- {citation_instruction}
+- Close with a brief sign-off and the placeholder "[Your Name]".
+- Plain text only. No markdown, no bullet points, no headers, no partisan framing.
+- Do not mention any political party.
+
+BILL: {bill_label} — {bill_title}
+STANCE: Vote {vote_word}
+{location_line}
+
+SELECTED POINTS TO REFERENCE:
+{points_block}
+
+Write the letter now:"""
+
+    return get_llm_provider(provider=llm_provider, model=llm_model).generate_text(prompt)