feat: rename keep-notes to memento-note, migrate to PostgreSQL, fix MCP bugs

- Rename directory keep-notes -> memento-note with all code references
- Prisma: SQLite -> PostgreSQL (both app and MCP server schemas)
- Sync MCP schema with main app (add missing fields, relations, indexes)
- Delete 17 SQLite migrations (clean slate for PostgreSQL)
- Remove SQLite dependencies (@libsql/client, better-sqlite3, etc.)
- Fix MCP server: hardcoded Windows DB paths -> DATABASE_URL env var
- Fix MCP server: .dockerignore excluded index-sse.js (SSE mode broken)
- MCP Dockerfile: node:20 -> node:22
- Docker Compose: add postgres service, remove SQLite volume
- Generate favicon.ico, icon-192.png, icon-512.png, apple-icon.png
- Update layout.tsx icons and manifest.json for PNG icons
- Update all .env files for PostgreSQL
- Rewrite README.md with updated sections
- Remove mcp-server/node_modules and prisma/client-generated from git tracking

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

memento-note/app/api/ai/models/route.ts

@@ -0,0 +1,96 @@
+import { NextRequest, NextResponse } from 'next/server'
+import { getSystemConfig } from '@/lib/config'
+
+// Modèles populaires pour chaque provider (2025)
+const PROVIDER_MODELS = {
+  ollama: {
+    tags: [
+      'llama3:latest',
+      'llama3.2:latest',
+      'granite4:latest',
+      'mistral:latest',
+      'mixtral:latest',
+      'phi3:latest',
+      'gemma2:latest',
+      'qwen2:latest'
+    ],
+    embeddings: [
+      'embeddinggemma:latest',
+      'mxbai-embed-large:latest',
+      'nomic-embed-text:latest'
+    ]
+  },
+  openai: {
+    tags: [
+      'gpt-4o',
+      'gpt-4o-mini',
+      'gpt-4-turbo',
+      'gpt-4',
+      'gpt-3.5-turbo'
+    ],
+    embeddings: [
+      'text-embedding-3-small',
+      'text-embedding-3-large',
+      'text-embedding-ada-002'
+    ]
+  },
+  custom: {
+    tags: [], // Will be loaded dynamically
+    embeddings: [] // Will be loaded dynamically
+  }
+}
+
+export async function GET(request: NextRequest) {
+  try {
+    const config = await getSystemConfig()
+    const provider = (config.AI_PROVIDER || 'ollama').toLowerCase()
+
+    let models = PROVIDER_MODELS[provider as keyof typeof PROVIDER_MODELS] || { tags: [], embeddings: [] }
+
+    // Pour Ollama, essayer de récupérer la liste réelle depuis l'API locale
+    if (provider === 'ollama') {
+      try {
+        const ollamaBaseUrl = config.OLLAMA_BASE_URL || process.env.OLLAMA_BASE_URL || 'http://localhost:11434'
+        const response = await fetch(`${ollamaBaseUrl}/api/tags`, {
+          method: 'GET',
+          headers: { 'Content-Type': 'application/json' }
+        })
+
+        if (response.ok) {
+          const data = await response.json()
+          const allModels = data.models || []
+
+          // Séparer les modèles de tags et d'embeddings
+          const tagModels = allModels
+            .filter((m: any) => !m.name.includes('embed') && !m.name.includes('Embedding'))
+            .map((m: any) => m.name)
+            .slice(0, 20) // Limiter à 20 modèles
+
+          const embeddingModels = allModels
+            .filter((m: any) => m.name.includes('embed') || m.name.includes('Embedding'))
+            .map((m: any) => m.name)
+
+          models = {
+            tags: tagModels.length > 0 ? tagModels : models.tags,
+            embeddings: embeddingModels.length > 0 ? embeddingModels : models.embeddings
+          }
+        }
+      } catch (error) {
+        // Garder les modèles par défaut
+      }
+    }
+
+    return NextResponse.json({
+      provider,
+      models: models || { tags: [], embeddings: [] }
+    })
+  } catch (error: any) {
+    return NextResponse.json(
+      {
+        error: error.message || 'Failed to fetch models',
+        models: { tags: [], embeddings: [] }
+      },
+      { status: 500 }
+    )
+  }
+}