[BUG] AbstractLogCollector.start() leaks thread pool and consumer thread on repeated calls

- severity: High
- files: `shenyu-plugin/shenyu-plugin-logging/shenyu-plugin-logging-common/src/main/java/org/apache/shenyu/plugin/logging/common/collector/AbstractLogCollector.java:72-83`
- description: `start()` creates a new `ShenyuThreadPoolExecutor` as a local variable (never stored as a field) and submits `this::consume` to it. No `started.compareAndSet` guard; does not shut down any previously created executor. `AbstractLogPluginDataHandler.handlerPlugin()` calls `logCollector().start()` on every config change. Each call leaks the old pool and starts a competing consumer on the same `bufferQueue`.
- impact: On every logging config refresh, a new thread pool and consumer leak; competing consumers cause duplicate log processing and thread exhaustion.
- suggested_fix: Store executor as a field; shut it down at the start of `start()`; guard with `started.compareAndSet(false, true)`.
- confidence: High
- related_existing: none

---
_Identified during the 2026-08-02 deep re-scan; full list in [`docs/scan2-2026-08-02/00-consolidated-critical-high.md`](docs/scan2-2026-08-02/00-consolidated-critical-high.md)._

Start with AbstractLogCollector.java:72-83 and trace the logCollector().start() call from AbstractLogPluginDataHandler.handlerPlugin(). Confirm how repeated configuration changes create competing consumers and retain executors. Done means repeated starts no longer leak pools or consumer threads, duplicate processing is prevented, and the existing logging behavior remains intact.

Written by the indexing model from the issue text.