[dify 1.15.0] Workflow execution is slower in this version than in the previous one.

### Self Checks

- [x] I have read the [Contributing Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) and [Language Policy](https://github.com/langgenius/dify/issues/1542).
- [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones.
- [x] I confirm that I am using English to submit this report, otherwise it will be closed.
- [x] 【中文用户 & Non English User】请使用英语提交，否则会被关闭 ：）
- [x] Please do not modify this template :) and fill in all the required fields.

### Dify version

1.15.0

### Cloud or Self Hosted

Self Hosted (Docker)

### Steps to reproduce

use workflows in version 1.15.0

### ✔️ Expected Behavior

run fast like before

### ❌ Actual Behavior

There are two separate performance regressions after the upgrade:

1. **A standalone workflow runs slower.**
The total execution time is now the sum of:

* the execution time of all internal nodes, **plus**
* an additional **~10 seconds of unexplained latency**.

This extra time does not appear in the execution trace of any node, but it is included in the workflow's total execution time.

2. **Calling another workflow via the Tool mechanism is also slower.**
Every time one workflow invokes another workflow as a Tool, there is an additional **~6 seconds of unexplained latency** before the called workflow executes. This delay is also not reflected in the node execution times.

Because these unexplained delays accumulate across multiple workflow calls, the overall performance degradation becomes very significant. A workflow that used to complete in **about 20 seconds** on the previous version now takes **around 50 seconds** on average after upgrading, even though the execution time of the individual nodes remains almost unchanged.

Start by reproducing workflow execution in the self-hosted Docker deployment on Dify 1.15.0, comparing it with the previous version. Inspect the workflow execution timing and Tool invocation path, using execution traces to isolate the unaccounted-for delays; done means the extra latency is explained and the upgraded version returns to the earlier performance.

Written by the indexing model from the issue text.