feat: add unified @manacore/shared-llm package and migrate all backends

Create a shared LLM client package that provides a unified interface
to the mana-llm service, replacing 9 individual fetch-based integrations
with consistent error handling, retry logic, and JSON extraction.

Package (@manacore/shared-llm):
- LlmModule with forRoot/forRootAsync (NestJS dynamic module)
- LlmClientService: chat, json, vision, visionJson, embed, stream
- LlmClient standalone class for non-NestJS consumers
- extractJson utility (consolidates 3 markdown-stripping implementations)
- retryFetch with exponential backoff (429, 5xx, network errors)
- 44 unit tests (json-extractor, retry, llm-client)

Migrated backends:
- mana-core-auth: raw fetch → llm.json()
- planta: raw fetch + vision → llm.visionJson()
- nutriphi: raw fetch + regex → llm.visionJson() + llm.json()
- chat: custom OllamaService (175 LOC) → llm.chatMessages()
- context: raw fetch → llm.chat() (keeps token tracking)
- traces: 2x raw fetch → llm.chat()
- manadeck: @google/genai SDK → llm.json() + llm.visionJson()
- bot-services: raw Ollama API → LlmClient standalone
- matrix-ollama-bot: raw fetch → llm.chatMessages() + llm.vision()

New credit operations:
- AI_PLANT_ANALYSIS (2 credits, planta)
- AI_GUIDE_GENERATION (5 credits, traces)
- AI_CONTEXT_GENERATION (2 credits, context)
- AI_BOT_CHAT (0.1 credits, matrix)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

packages/shared-llm/src/interfaces/index.ts

@@ -0,0 +1,8 @@
+export type {
+	LlmModuleOptions,
+	LlmModuleAsyncOptions,
+	LlmOptionsFactory,
+	ResolvedLlmOptions,
+} from './llm-options.interface';
+
+export { resolveOptions } from './llm-options.interface';

packages/shared-llm/src/interfaces/llm-options.interface.ts

@@ -0,0 +1,47 @@
+import type { ModuleMetadata, Type } from '@nestjs/common';
+
+export interface LlmModuleOptions {
+	/** mana-llm service URL (default: http://localhost:3025) */
+	manaLlmUrl?: string;
+	/** Default text model (default: ollama/gemma3:4b) */
+	defaultModel?: string;
+	/** Default vision model (default: ollama/llava:7b) */
+	defaultVisionModel?: string;
+	/** Request timeout in ms (default: 120000) */
+	timeout?: number;
+	/** Max retries on transient failures (default: 2) */
+	maxRetries?: number;
+	/** Enable debug logging (default: false) */
+	debug?: boolean;
+}
+
+export interface LlmModuleAsyncOptions extends Pick<ModuleMetadata, 'imports'> {
+	useExisting?: Type<LlmOptionsFactory>;
+	useClass?: Type<LlmOptionsFactory>;
+	useFactory?: (...args: any[]) => Promise<LlmModuleOptions> | LlmModuleOptions;
+	inject?: any[];
+}
+
+export interface LlmOptionsFactory {
+	createLlmOptions(): Promise<LlmModuleOptions> | LlmModuleOptions;
+}
+
+export interface ResolvedLlmOptions {
+	manaLlmUrl: string;
+	defaultModel: string;
+	defaultVisionModel: string;
+	timeout: number;
+	maxRetries: number;
+	debug: boolean;
+}
+
+export function resolveOptions(options: LlmModuleOptions): ResolvedLlmOptions {
+	return {
+		manaLlmUrl: options.manaLlmUrl ?? 'http://localhost:3025',
+		defaultModel: options.defaultModel ?? 'ollama/gemma3:4b',
+		defaultVisionModel: options.defaultVisionModel ?? 'ollama/llava:7b',
+		timeout: options.timeout ?? 120_000,
+		maxRetries: options.maxRetries ?? 2,
+		debug: options.debug ?? false,
+	};
+}