Optimize issue_graph crawler performance

- Add priority-based concurrent crawling with heap queue
- Parents get highest priority, children next, cross-refs lowest
- Timeline only fetched for focus node (reduces API calls)
- Cross-references limited to one hop (prevents explosion)
- Add isAncestor flag to skip sibling crawling
- Add verbose mode with crawl statistics
- Add tasklist extraction for batch/epic issues
- Reduce MaxGraphDepth to 4, MaxConcurrentFetches to 5
- Add rate limit backoff with exponential retry
- Remove unused fields from crawlItem/crawlResult structs
- Simplify heap implementation (no index tracking needed)

Results: 13 nodes in <10s (was timing out), focused on direct lineage

Author: SamMorrowDrums

README.md

@@ -722,6 +722,7 @@ The following sets of tools are available:
   - `issue_number`: Issue or pull request number to build the graph from (number, required)
   - `owner`: Repository owner (string, required)
   - `repo`: Repository name (string, required)
+  - `verbose`: Include crawl statistics showing how the graph was traversed (nodes fetched, depth reached, repos accessed, etc.) (boolean, optional)
 
 - **issue_read** - Get issue details
   - `issue_number`: The number of the issue (number, required)

pkg/github/__toolsnaps__/issue_graph.snap

@@ -26,6 +26,10 @@
       "repo": {
         "description": "Repository name",
         "type": "string"
+      },
+      "verbose": {
+        "description": "Include crawl statistics showing how the graph was traversed (nodes fetched, depth reached, repos accessed, etc.)",
+        "type": "boolean"
       }
     },
     "required": [