docs(release-notes): update v1.81.14 key highlights and section placement

docs/my-website/release_notes/v1.81.14.md

@@ -43,10 +43,10 @@ pip install litellm==1.81.14
 ## Key Highlights
 
 - **Use Claude Sonnet 4.6 on day 0** — [reasoning, computer use, prompt caching, and 200K context, working across Anthropic and Vertex AI from the moment it launched](../../docs/providers/anthropic)
-- **Deploy guardrails without writing code** — [Guardrail Garden lets you browse a marketplace of pre-built policies (competitor blockers, GDPR PII, EU AI Act, prompt injection) and deploy in one click](../../docs/proxy/guardrails/policy_templates)
+- **Guardrail Model Garden** — [Easily browse available guardrails — competitor blockers, GDPR PII, EU AI Act, prompt injection, and more — and deploy in one click](../../docs/proxy/guardrails/policy_templates)
+- **3 new built-in LiteLLM guardrails** — [competitor name blocker, topic blocker, and insults filter — all configurable, no external API required](../../docs/proxy/guardrails)
 - **Test guardrail policies before shipping** — [upload a CSV dataset to the compliance playground and validate policies against real traffic; get AI-generated policy suggestions with latency overhead estimates](../../docs/proxy/guardrails/policy_templates)
 - **Turn any OpenAPI spec into an MCP server** — [paste a spec and get a working MCP server instantly, via API or UI](../../docs/mcp)
-- **Call any prompt management system from a single API** — [the new Prompt Management API works with Langfuse, LangSmith, and others without requiring per-integration code](../../docs/proxy/prompt_management)
 - **Major performance batch** — 20+ targeted optimizations across router algorithms, logging overhead, cost calculator, and connection management — meaningfully lower latency and CPU overhead on every request
 
 ---
@@ -108,7 +108,6 @@ This release includes the largest single batch of performance work since v1.74.
     - Native structured outputs API support (`outputConfig.textFormat`) - [PR #21222](https://github.com/BerriAI/litellm/pull/21222)
     - Support `nova/` and `nova-2/` spec prefixes for custom imported models - [PR #21359](https://github.com/BerriAI/litellm/pull/21359)
     - Broaden Nova 2 model detection to support all `nova-2-*` variants - [PR #21358](https://github.com/BerriAI/litellm/pull/21358)
-    - Add Accept header for AgentCore MCP server requests - [PR #21551](https://github.com/BerriAI/litellm/pull/21551)
     - Clamp `thinking.budget_tokens` to minimum 1024 - [PR #21306](https://github.com/BerriAI/litellm/pull/21306)
     - Fix `parallel_tool_calls` mapping for Bedrock Converse - [PR #21659](https://github.com/BerriAI/litellm/pull/21659)
 
@@ -330,6 +329,10 @@ This release includes the largest single batch of performance work since v1.74.
 - **[LakeraAI](../../docs/proxy/guardrails)**
     - Avoid `KeyError` on missing `LAKERA_API_KEY` during initialization - [PR #21422](https://github.com/BerriAI/litellm/pull/21422)
 
+### Auto Routing
+
+- **Complexity-based auto routing** — new router strategy that scores requests across 7 dimensions (token count, code presence, reasoning markers, technical terms, etc.) and routes to the appropriate model tier — no embeddings or API calls required - [PR #21789](https://github.com/BerriAI/litellm/pull/21789), [Docs](../../docs/proxy/auto_routing)
+
 ### Prompt Management
 
 - **Prompt Management API**
@@ -386,7 +389,6 @@ This release includes the largest single batch of performance work since v1.74.
 - Increase default LRU cache size to reduce multi-model cache thrash - [PR #21139](https://github.com/BerriAI/litellm/pull/21139)
 - Cache `get_model_access_groups()` no-args result on Router - [PR #20374](https://github.com/BerriAI/litellm/pull/20374)
 - Deployment affinity routing callback — route to the same deployment for a session - [PR #19143](https://github.com/BerriAI/litellm/pull/19143)
-- Complexity-based auto routing — new router strategy that routes based on request complexity - [PR #21789](https://github.com/BerriAI/litellm/pull/21789)
 - Session-ID-based routing — use `session_id` for consistent routing within a session - [PR #21763](https://github.com/BerriAI/litellm/pull/21763)
 
 **Connection management & reliability**