feat: phase 2 of AI pipeline hardening — tool-based structured outputs + prompt caching

Every structured-output call now uses an Anthropic tool instead of
parsing JSON out of free-form prose, and stable system prompts are
sent as cacheable blocks. Behaviour-equivalent to phase 1 from the
caller's point of view; the savings show up in token usage and in the
absence of "AI returned non-JSON response" failure modes.

* src/lib/llmTools.js — single source of truth for tool definitions:
  emit_knowledge_graph, emit_handbook_delta, emit_learning_article /
  _slides / _infographic / _all, emit_custom_topic, emit_quiz_questions,
  emit_graph_actions, plus five article-patch tools (set_intro,
  set_section, add_section, remove_section, replace_takeaways).
* src/lib/articlePatches.js — pure applyArticlePatches +
  applyAndValidate; rebuilds the article from a sequence of patch tool
  calls and re-validates against learningArticleSchema. set_section
  falls back to appending when no matching heading exists so the
  model's intent is preserved rather than silently dropped.
* src/lib/llmSchemas.js — Zod schemas for the five patch ops,
  registered in toolSchemaRegistry so callLLM validates them
  automatically.
* src/lib/llm.js — simulation mode now returns a tool_use stub matching
  toolChoice.name, so the UI keeps working with Simulation Mode on
  after the structured-output migration.
* src/lib/extractionPipeline.js — processSourceText and
  analyzeHandbookDelta migrated to callLLM + tool use. System prompts
  sent as { cache_control: ephemeral } blocks. Handbook results pass
  through normalizeHandbookResult to collapse legacy "executes"
  relations into executed_by with swapped source/target.
* src/lib/learningService.js — generateLearningContent picks the right
  tool per selectedType; generateCustomTopic uses emit_custom_topic;
  refineLearningContent now drives the five patch tools with
  toolChoice 'any' and rejects the whole turn if the patched article
  fails validation. Article-only refinement is intentional for phase 2;
  refining a topic without an article surfaces a clear error.
* src/lib/testService.js — quiz generation via emit_quiz_questions.
* src/components/admin/KnowledgeGraph.jsx — analyzeGraph routed through
  the reasoning tier (Opus) since graph-wide consolidation benefits
  from a stronger reasoner.
* src/components/chat/prompts.js — buildSystemPrompt now returns three
  text blocks: stable preamble (cached), KB context (cached, hash-bust
  deferred to phase 5), per-turn user/admin tail (uncached).
* src/lib/__tests__/ — 13 new tests covering each patch op, multi-op
  sequencing, post-patch validation failure, and tool/registry shape.

Acceptance: lint and 45/45 tests green; build succeeds; no
`match(/\{[\s\S]*\}/)` JSON extraction left in src/. Live verification
of cache hits on a second extraction within 5 minutes is deferred to
manual smoke testing — needs real `/api/anthropic` traffic.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/lib/llmSchemas.js

@@ -183,6 +183,31 @@ export const proposeGraphDeltaSchema = z.object({
   relations: z.array(deltaRelationSchema).max(5).optional(),
 });
 
+// ── Article patch operation schemas (Phase 2.4) ──────────────────────────────
+
+export const setIntroPatchSchema = z.object({
+  intro: z.string().min(1),
+});
+
+export const setSectionPatchSchema = z.object({
+  heading: z.string().min(1),
+  body: z.string().min(1),
+});
+
+export const addSectionPatchSchema = z.object({
+  heading: z.string().min(1),
+  body: z.string().min(1),
+  position: z.enum(['start', 'end']),
+});
+
+export const removeSectionPatchSchema = z.object({
+  heading: z.string().min(1),
+});
+
+export const replaceTakeawaysPatchSchema = z.object({
+  items: z.array(z.string().min(1)).min(1),
+});
+
 /**
  * Registry mapping known tool names to their input schemas. `callLLM`
  * consults this when the caller does not pass an explicit `toolSchemas`
@@ -199,4 +224,9 @@ export const toolSchemaRegistry = {
   emit_custom_topic: customTopicSchema,
   emit_graph_actions: graphActionsSchema,
   propose_graph_delta: proposeGraphDeltaSchema,
+  set_intro: setIntroPatchSchema,
+  set_section: setSectionPatchSchema,
+  add_section: addSectionPatchSchema,
+  remove_section: removeSectionPatchSchema,
+  replace_takeaways: replaceTakeawaysPatchSchema,
 };