fix(sovereignty): complete Anthropic purge — the-nexus (closes #1143)

Mirrors timmy-config PR #440 (Anthropic purge complete).
Re-applies PR #1394 changes to current main (PR was closed unmerged).

== Core purge (16 files — PR #1394) ==
- Frontend: app.js, style.css, portal & world state — all CLAUDE → GEMINI
- Fleet: ezra model google/gemini-2.5-pro; hermes-trismegistus lane deprecated
- Config: deepdive env/examples — Anthropic removed, default to Gemini
- Docs: FLEET_VOCABULARY fallback chain updated (Kimi > Gemini > Ollama)
- Components: agent presence & heartbeat panels — CLAUDE → GEMINI standby

== Ancillary fixes (3 files — active code references) ==
- bin/deepdive_filter.py — annotate "anthropic/claude" as research keywords (competitor monitoring)
- scaffold/deep-dive/aggregator/arxiv_fetcher.py — annotate RELEVANCE_KEYWORDS "claude" as research keyword
- intelligence/deepdive/prompts/production_briefing_v1.txt — genericize conversational-source example

== Preserved (per issue scope) ==
Historical (docs, reports, audits), tests, CLAUDE.md, .gitattributes,
.git hooks, competitor-monitoring configs (sources, keywords, RSS feeds),
architecture linter, sovereignty tests, deepdive historical metadata.

No tests touched; no historical documents modified.

config/deepdive.env.example

@@ -8,12 +8,9 @@
 # Primary: OpenRouter (recommended - access to multiple models)
 OPENROUTER_API_KEY=sk-or-v1-...
 DEEPDIVE_LLM_PROVIDER=openrouter
-DEEPDIVE_LLM_MODEL=anthropic/claude-sonnet-4
+DEEPDIVE_LLM_MODEL=google/gemini-2.5-pro
 
-# Alternative: Anthropic direct
-# ANTHROPIC_API_KEY=sk-ant-...
-# DEEPDIVE_LLM_PROVIDER=anthropic
-# DEEPDIVE_LLM_MODEL=claude-3-5-sonnet-20241022
+# Anthropic purged — Kimi/Gemini/Ollama only
 
 # Alternative: OpenAI
 # OPENAI_API_KEY=sk-...