docs: comprehensive docs audit — cover 13 features from last week's PRs (#5815)

Cover documentation gaps found by auditing all 50+ merged PRs from the past week:

tools-reference.md:
- Fix stale tool count (47→46, 11→10 browser tools) after browser_close removal
- Document notify_on_complete parameter in terminal tool description

telegram.md:
- Add Interactive Model Picker section (inline keyboard, provider/model drill-down)

discord.md:
- Add Interactive Model Picker section (Select dropdowns, 120s timeout)
- Add Native Slash Commands for Skills section (auto-registration at startup)

signal.md:
- Expand Attachments section with outgoing media delivery (send_image_file,
  send_voice, send_video, send_document via MEDIA: tags)

webhooks.md:
- Document {__raw__} special template token for full payload access
- Document Forum Topic Delivery via message_thread_id in deliver_extra

slack.md:
- Fix stale/misleading thread reply docs — thread replies no longer require
  @mention when bot has active session (3 locations updated)

security.md:
- Add cross-session isolation (layer 6) and input sanitization (layer 7)
  to security layers overview

feishu.md:
- Add WebSocket Tuning section (ws_reconnect_interval, ws_ping_interval)
- Add Per-Group Access Control section (group_rules with 5 policy types)

credential-pools.md:
- Add Delegation & Subagent Sharing section

delegation.md:
- Update key properties to mention credential pool inheritance

providers.md:
- Add Z.AI Endpoint Auto-Detection note
- Add xAI (Grok) Prompt Caching section

skills-catalog.md:
- Add p5js to creative skills category

website/docs/integrations/providers.md

@@ -168,6 +168,16 @@ model:
 
 Base URLs can be overridden with `GLM_BASE_URL`, `KIMI_BASE_URL`, `MINIMAX_BASE_URL`, `MINIMAX_CN_BASE_URL`, or `DASHSCOPE_BASE_URL` environment variables.
 
+:::note Z.AI Endpoint Auto-Detection
+When using the Z.AI / GLM provider, Hermes automatically probes multiple endpoints (global, China, coding variants) to find one that accepts your API key. You don't need to set `GLM_BASE_URL` manually — the working endpoint is detected and cached automatically.
+:::
+
+### xAI (Grok) Prompt Caching
+
+When using xAI as a provider (any base URL containing `x.ai`), Hermes automatically enables prompt caching by sending the `x-grok-conv-id` header with every API request. This routes requests to the same server within a conversation session, allowing xAI's infrastructure to reuse cached system prompts and conversation history.
+
+No configuration is needed — caching activates automatically when an xAI endpoint is detected and a session ID is available. This reduces latency and cost for multi-turn conversations.
+
 ### Hugging Face Inference Providers
 
 [Hugging Face Inference Providers](https://huggingface.co/docs/inference-providers) routes to 20+ open models through a unified OpenAI-compatible endpoint (`router.huggingface.co/v1`). Requests are automatically routed to the fastest available backend (Groq, Together, SambaNova, etc.) with automatic failover.