feat: attention-optimized semantic protocol v2.7

Core changes:
- Add @defgroup/@ingroup to 1791 C2+ contracts (555 files) for HCA 128× pre-training DSA grouping
- Add §0.1 Pre-Training Frequency matrix to semantics-core
- Add §VIII Attention Architecture rules (ATTN_1-4) with MLA/CSA/HCA/DSA mechanics
- Add @defgroup/@ingroup to canonical syntax (§II) and all contract examples

Agent prompts (5 files):
- Add ZERO-STATE RATIONALE with MLA/CSA/HCA/DSA compression mechanics
- Add pre-training note: @RATIONALE/@REJECTED are in-context learned tags
- svelte-coder: add missing #region contract, fix Svelte rule violations
- python-coder/fullstack-coder: honor function contracts from speckit plan
- qa-tester: add attention compliance audit (P3 ATTN_1-4 checks)

Skills (6 files):
- Translate all axiom_config descriptions to English
- Fix doc_dirs to index .opencode/ and .specify/
- Deduplicate 5× complexity_rules → single global_tags catalog
- Reduce semantics-svelte 591→485 lines (remove duplicate code blocks)
- Fix semantics-testing: 'Short IDs' → 'Short hierarchical IDs'
- Fix all examples: flat IDs → hierarchical Domain.Name format
- Fix Svelte examples: replace raw Tailwind + <button> with semantic tokens + /ui

Speckit workflow (commands + templates):
- speckit.plan: add Function-Level Contracts for C3+ with @PRE/@POST/@TEST_EDGE
- speckit.plan: add Attention Compliance Gate (ATTN_1-4 before contract generation)
- speckit.tasks: add function contract inlining format (constraints in task description)
- speckit.specify: load semantics-core for spec density rules
- spec-template: add #region contract, @SEMANTICS grouping, hierarchical IDs
- ux-reference-template: add #region wrapper
- plan-template: add attention gate, @defgroup/@ingroup guidance
- tasks-template: add attention audit + rebuild + orphan check tasks
- constitution.md: translate to English, add Principle VIII (attention-optimized contracts)

Reference modules rewritten (hierarchical IDs + full contracts):
- Auth.Jwt: 6 child contracts with @RATIONALE/@REJECTED/@TEST_EDGE
- Api.Auth: 5 endpoints with @TEST_EDGE + molecular CoT markers
- Migration.Model: @defgroup Migration with 18 @ACTION + 6 @INVARIANT

Scripts:
- add_defgroup_ingroup.py: zero-risk additive @ingroup migration (1791 insertions)
- migrate_hierarchical.py: flat→hierarchical ID dry-run analysis (792 contracts)
- merge_prompts.py: merge all prompts/skills/commands into one review file

Config:
- axiom_config.yaml: 749→395 lines (-47%), English, doc_dirs include prompts
- Fix test_datasets.py import collision (rename → test_datasets_routes.py)
- Fix test_preview.py: SupersetClient→get_superset_client, AsyncMock, logger f-string

backend/src/plugins/llm_analysis/__init__.py

@@ -1,4 +1,5 @@
 # #region LLMAnalysisPackage [TYPE Module] [SEMANTICS llm, analysis, plugin, package, export]
+# @defgroup LLMAnalysis Module group.
 
 """
 LLM Analysis Plugin for automated dashboard validation and dataset documentation.

backend/src/plugins/llm_analysis/_topology.py

@@ -1,4 +1,5 @@
 # #region DashboardTopology [C:3] [TYPE Module] [SEMANTICS superset, topology, dashboard, layout, parsing]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Dashboard topology extraction — parses Superset position_json into human-readable
 #   text descriptions of Tab → Row → Chart hierarchy for LLM prompt context.
 # @LAYER Plugin
@@ -227,6 +228,7 @@ def _extract_topology_children(
 
 
 # #region build_dashboard_topology [C:2] [TYPE Function]
+# @ingroup LLMAnalysis
 # @BRIEF Build a structured text description of dashboard layout from position_json.
 # @PARAM  dashboard (dict) - Dashboard metadata dict with position_json, dashboard_title.
 # @PARAM  charts (list[dict]) - List of fetched chart detail dicts for name/viz lookup.

backend/src/plugins/llm_analysis/migrations/v1_to_v2.py

@@ -1,4 +1,5 @@
 # #region V1ToV2Migration [C:3] [TYPE Module] [SEMANTICS migration,validation,data,idempotent]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Migration script: v1 → v2 ValidationPolicy data. Converts existing dashboard_ids
 #           JSON arrays to ValidationSource relational rows.
 # @PRE  Database has both validation_policies and validation_sources tables.

backend/src/plugins/llm_analysis/models.py

@@ -1,4 +1,5 @@
 # #region LLMAnalysisModels [C:3] [TYPE Module] [SEMANTICS pydantic, llm, plugin, model, provider-type]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Define Pydantic models for LLM Analysis plugin.
 # @LAYER Domain
 # @RELATION DEPENDS_ON -> [EXT:Library:pydantic]
@@ -10,6 +11,7 @@ from pydantic import BaseModel, Field
 
 
 # #region LLMProviderType [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Enum for supported LLM providers.
 class LLMProviderType(str, Enum):
     OPENAI = "openai"
@@ -19,6 +21,7 @@ class LLMProviderType(str, Enum):
 # #endregion LLMProviderType
 
 # #region LLMProviderConfig [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Configuration for an LLM provider.
 class LLMProviderConfig(BaseModel):
     id: str | None = None
@@ -41,6 +44,7 @@ class LLMProviderConfig(BaseModel):
 # #endregion LLMProviderConfig
 
 # #region ValidationStatus [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Enum for dashboard validation status.
 class ValidationStatus(str, Enum):
     PASS = "PASS"
@@ -50,6 +54,7 @@ class ValidationStatus(str, Enum):
 # #endregion ValidationStatus
 
 # #region DetectedIssue [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Model for a single issue detected during validation.
 class DetectedIssue(BaseModel):
     severity: ValidationStatus
@@ -58,6 +63,7 @@ class DetectedIssue(BaseModel):
 # #endregion DetectedIssue
 
 # #region ValidationResult [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Model for dashboard validation result.
 class ValidationResult(BaseModel):
     id: str | None = None

backend/src/plugins/llm_analysis/plugin.py

@@ -1,4 +1,5 @@
 # #region LLMAnalysisPlugin [C:5] [TYPE Module] [SEMANTICS llm, analysis, dashboard, validation, documentation]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Implements DashboardValidationPlugin and DocumentationPlugin.
 # @LAYER Plugin
 # @RELATION INHERITS -> [PluginBase]
@@ -159,6 +160,7 @@ def _update_run_status(db, run_id: str | None) -> None:
 
 
 # #region DashboardValidationPlugin [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Plugin for automated dashboard health analysis using LLMs.
 # @RELATION IMPLEMENTS -> [PluginBase]
 class DashboardValidationPlugin(PluginBase):
@@ -846,6 +848,7 @@ class DashboardValidationPlugin(PluginBase):
 
 
 # #region DocumentationPlugin [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Plugin for automated dataset documentation using LLMs.
 # @RELATION IMPLEMENTS -> [PluginBase]
 class DocumentationPlugin(PluginBase):

backend/src/plugins/llm_analysis/scheduler.py

@@ -1,4 +1,5 @@
 # #region LLMAnalysisScheduler [C:3] [TYPE Module] [SEMANTICS scheduler, llm, validation, task, cron]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Provides helper functions to schedule LLM-based validation tasks.
 # @LAYER Domain
 # @RELATION DEPENDS_ON -> [SchedulerService]
@@ -10,6 +11,7 @@ from ...dependencies import get_scheduler_service, get_task_manager
 
 
 # #region schedule_dashboard_validation [TYPE Function]
+# @ingroup LLMAnalysis
 # @BRIEF Schedules a recurring dashboard validation task.
 # @SIDE_EFFECT  Adds a job to the scheduler service.
 def schedule_dashboard_validation(dashboard_id: str, cron_expression: str, params: dict[str, Any]):

backend/src/plugins/llm_analysis/service.py

@@ -1,4 +1,5 @@
 # #region LLMAnalysisService [C:5] [TYPE Module] [SEMANTICS llm, screenshot, playwright, openai, tenacity]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Services for LLM interaction and dashboard screenshots.
 # @LAYER Plugin
 # @RELATION DEPENDS_ON -> [EXT:Library:tenacity]
@@ -37,6 +38,7 @@ LLM_HTTP_TIMEOUT_S = 120  # seconds (httpx client timeout)
 DEFAULT_USER_AGENT = "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36"
 
 # #region ScreenshotService [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Handles capturing screenshots of Superset dashboards.
 class ScreenshotService:
     # region ScreenshotService.__init__ [TYPE Function]
@@ -837,6 +839,7 @@ class ScreenshotService:
 # #endregion ScreenshotService
 
 # #region LLMClient [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Wrapper for LLM provider APIs.
 class LLMClient:
     # region LLMClient.__init__ [TYPE Function]
@@ -1441,6 +1444,7 @@ class LLMClient:
 # #endregion LLMClient
 
 # #region DatasetHealthChecker [C:3] [TYPE Class]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Checks dataset accessibility and KXD connectivity via Superset API.
 # @LAYER Service
 # @RELATION CALLS -> [SupersetClient]
@@ -1641,6 +1645,7 @@ class DatasetHealthChecker:
 # #endregion DatasetHealthChecker
 
 # #region RedactionService [C:2] [TYPE Module]
+# @defgroup LLMAnalysis Module group.
 # @BRIEF Redacts PII, credentials, and sensitive data from logs and LLM responses.
 # @LAYER Service
 # @RATIONALE FR-029: sensitive data must be filtered BEFORE external LLM send and BEFORE persistence.