feat(sync): fetch and store GitLab issue links (bd-343o)

Add end-to-end support for GitLab issue link fetching:
- New GitLabIssueLink type + fetch_issue_links API client method
- Migration 029: add issue_links job type and watermark column
- issue_links.rs: bidirectional entity_reference storage with
  self-link skip, cross-project fallback, idempotent upsert
- Drain pipeline in orchestrator following mr_closes_issues pattern
- Display related issues in 'lore show issues' output
- --no-issue-links CLI flag with config, autocorrect, robot-docs
- 6 unit tests for storage logic

migrations/029_issue_links_job_type.sql

@@ -0,0 +1,43 @@
+-- Migration 029: Expand pending_dependent_fetches CHECK to include 'issue_links' job type.
+-- Also adds issue_links_synced_for_updated_at watermark to issues table.
+-- SQLite cannot ALTER CHECK constraints, so we recreate the table.
+
+-- Step 1: Recreate pending_dependent_fetches with expanded CHECK
+CREATE TABLE pending_dependent_fetches_new (
+    id INTEGER PRIMARY KEY,
+    project_id INTEGER NOT NULL REFERENCES projects(id) ON DELETE CASCADE,
+    entity_type TEXT NOT NULL CHECK (entity_type IN ('issue', 'merge_request')),
+    entity_iid INTEGER NOT NULL,
+    entity_local_id INTEGER NOT NULL,
+    job_type TEXT NOT NULL CHECK (job_type IN (
+        'resource_events', 'mr_closes_issues', 'mr_diffs', 'issue_links'
+    )),
+    payload_json TEXT,
+    enqueued_at INTEGER NOT NULL,
+    locked_at INTEGER,
+    attempts INTEGER NOT NULL DEFAULT 0,
+    next_retry_at INTEGER,
+    last_error TEXT
+);
+
+INSERT INTO pending_dependent_fetches_new
+    SELECT * FROM pending_dependent_fetches;
+
+DROP TABLE pending_dependent_fetches;
+
+ALTER TABLE pending_dependent_fetches_new RENAME TO pending_dependent_fetches;
+
+-- Recreate indexes from migration 011
+CREATE UNIQUE INDEX uq_pending_fetches
+    ON pending_dependent_fetches(project_id, entity_type, entity_iid, job_type);
+CREATE INDEX idx_pending_fetches_claimable
+    ON pending_dependent_fetches(job_type, locked_at) WHERE locked_at IS NULL;
+CREATE INDEX idx_pending_fetches_retryable
+    ON pending_dependent_fetches(next_retry_at) WHERE locked_at IS NULL AND next_retry_at IS NOT NULL;
+
+-- Step 2: Add watermark column for issue link sync tracking
+ALTER TABLE issues ADD COLUMN issue_links_synced_for_updated_at INTEGER;
+
+-- Update schema version
+INSERT INTO schema_version (version, applied_at, description)
+VALUES (29, strftime('%s', 'now') * 1000, 'Expand dependent fetch queue for issue links');