feat: multi-provider AI support with per-provider model selection

- Add llm-factory.ts: unified LLM provider abstraction (Google/Claude/OpenAI)
- Install @langchain/anthropic and @langchain/openai packages
- resolveAiProvider picks active provider from DB settings (Claude > OpenAI > Google)
- All AI graphs (reminder, new-patient, reschedule, internal-chat) now accept provider+model params
- Add claudeAiModel, openAiModel, googleAiModel columns to ai_settings table
- New PUT /api/ai/provider-model route to save selected model per provider
- UI model dropdowns for Claude (Haiku/Sonnet/Opus), OpenAI (GPT-5.x series), Google (Gemini 2.5/3.x)
- Google AI section also gets model selector alongside existing API key field

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

apps/Backend/src/ai/reminder-graph.ts

@@ -1,5 +1,5 @@
 import { StateGraph, END, START, Annotation } from "@langchain/langgraph";
-import { ChatGoogleGenerativeAI } from "@langchain/google-genai";
+import { getLlm, type AiProvider } from "./llm-factory";
 
 const GraphState = Annotation.Root({
   message:             Annotation<string>(),
@@ -95,7 +95,9 @@ async function confirmNode(state: GraphStateType, config: any) {
   if (!apiKey) return { reply: fallback };
 
   try {
-    const llm = new ChatGoogleGenerativeAI({ model: "gemini-1.5-flash", apiKey });
+    const provider: AiProvider = config?.configurable?.provider ?? "google";
+    const model: string | undefined = config?.configurable?.model;
+    const llm = getLlm(provider, apiKey, model);
     const apptClause = appt ? ` Their appointment is on ${appt}.` : "";
     const response = await llm.invoke([
       {
@@ -134,7 +136,9 @@ async function rescheduleNode(state: GraphStateType, config: any) {
   if (!apiKey) return { reply: fallback };
 
   try {
-    const llm = new ChatGoogleGenerativeAI({ model: "gemini-1.5-flash", apiKey });
+    const provider: AiProvider = config?.configurable?.provider ?? "google";
+    const model: string | undefined = config?.configurable?.model;
+    const llm = getLlm(provider, apiKey, model);
     const response = await llm.invoke([
       {
         role: "system",
@@ -160,7 +164,9 @@ async function otherNode(state: GraphStateType, config: any) {
     const fallback = NEW_APPT_FALLBACKS[lang] ?? NEW_APPT_FALLBACKS["English"]!;
     if (!apiKey) return { reply: fallback, intent: "wants_appointment" };
     try {
-      const llm = new ChatGoogleGenerativeAI({ model: "gemini-1.5-flash", apiKey });
+      const provider: AiProvider = config?.configurable?.provider ?? "google";
+    const model: string | undefined = config?.configurable?.model;
+    const llm = getLlm(provider, apiKey, model);
       const response = await llm.invoke([
         {
           role: "system",
@@ -177,7 +183,9 @@ async function otherNode(state: GraphStateType, config: any) {
   const fallback = state.generalFallback || (GENERAL_FALLBACKS[lang] ?? GENERAL_FALLBACKS["English"]!);
   if (!apiKey) return { reply: fallback };
   try {
-    const llm = new ChatGoogleGenerativeAI({ model: "gemini-1.5-flash", apiKey });
+    const provider: AiProvider = config?.configurable?.provider ?? "google";
+    const model: string | undefined = config?.configurable?.model;
+    const llm = getLlm(provider, apiKey, model);
     const response = await llm.invoke([
       {
         role: "system",
@@ -215,11 +223,13 @@ export async function runReminderGraph(
   language            = "English",
   appointmentDatetime = "",
   rescheduleGreeting  = "",
-  generalFallback     = ""
+  generalFallback     = "",
+  provider: AiProvider = "google",
+  model?: string
 ): Promise<{ reply: string | null; intent: string | null }> {
   const result = await graph.invoke(
     { message: patientMessage, intent: "", reply: "", language, appointmentDatetime, rescheduleGreeting, generalFallback },
-    { configurable: { apiKey } }
+    { configurable: { apiKey, provider, model } }
   );
   return {
     reply:  result.reply  || null,