feat: 新增Worker标签与DAG槽位显示，优化部署配置

如果您需要在真实的集群（1个 Master 节点 + 多个 Worker 节点）上部署该系统，我们提供了开箱即用的 Ansible Playbooks。由于我们有多个不同的部署环境（如 `p368`, `csu`, `zjlab`），我们为每个环境准备了专属的 Ansible Inventory 文件。

**💡 进阶：自定义 Worker 节点的并发度**

您可以在 Inventory 文件（例如 `inventory.p368.ini`）中，为每台 Worker 机器灵活配置变量：

- `worker_concurrency`: 设置该台机器上 Worker 容器的 Slots 数量（例如 16 或 64）。

*(通过这种方式，您可以让高配物理机加大并发，低配物理机少跑，充分压榨集群性能！)*

**💡 进阶：为每个 Worker 单独指定并发度**

您可以直接在 Inventory 文件（例如 `inventory.p368.ini`）的每个 Worker 节点条目上设置 `worker_concurrency`，从而让不同机器使用不同的 Celery 并发数：

```ini

[worker]

worker-a ansible_host=192.168.25.19 ansible_user=root deploy_dir=/opt/csst-airflow-worker-a worker_concurrency=16

worker-b ansible_host=192.168.25.20 ansible_user=root deploy_dir=/opt/csst-airflow-worker-b worker_concurrency=64

worker-c ansible_host=192.168.25.21 ansible_user=root deploy_dir=/opt/csst-airflow-worker-c worker_concurrency=8

```

- `worker_concurrency`：该节点上 Airflow Celery Worker 的并发槽位数。

- 建议按机器 CPU / 内存资源分别设置，不要所有 Worker 都使用同一个值。

- 如无特殊需要，请在每个 Worker 上显式写出该值，避免依赖默认值。

#### 1. 准备工作

- **安装 Ansible**：如果您的执行机（通常是您的本地电脑或 Master 节点）尚未安装 Ansible，请先通过以下命令安装：

node-master ansible_host=192.168.25.18 ansible_user=root deploy_dir=/opt/csst-airflow-master

[worker]

node-worker1 ansible_host=192.168.25.19 ansible_user=root deploy_dir=/opt/csst-airflow-worker1

node-worker2 ansible_host=192.168.25.20 ansible_user=root deploy_dir=/opt/csst-airflow-worker2

node-worker1 ansible_host=192.168.25.19 ansible_user=root deploy_dir=/opt/csst-airflow-worker1 worker_concurrency=16

node-worker2 ansible_host=192.168.25.20 ansible_user=root deploy_dir=/opt/csst-airflow-worker2 worker_concurrency=32

[all:vars]

env_name=csu

deploy_dir=/opt/csst-airflow

```

如果多个 Worker 机器规格相同，也可以在 `[worker:vars]` 中统一配置默认值，再在个别节点上单独覆盖：

```ini

[worker:vars]

worker_concurrency=16

[worker]

node-worker1 ansible_host=192.168.25.19 ansible_user=root deploy_dir=/opt/csst-airflow-worker1

node-worker2 ansible_host=192.168.25.20 ansible_user=root deploy_dir=/opt/csst-airflow-worker2

node-worker3 ansible_host=192.168.25.21 ansible_user=root deploy_dir=/opt/csst-airflow-worker3 worker_concurrency=32

```

上述配置会在部署时写入目标节点的 `.env`，并在启动 Worker 容器时传递给 `AIRFLOW__CELERY__WORKER_CONCURRENCY`，因此只需要修改 Inventory 后重新执行 Ansible 部署或更新即可生效。

#### 3. 配置环境变量

为了让集群中的所有组件（特别是 Worker 节点）能够正确找到主节点和其他依赖，您需要为该环境配置专用的 `.env` 文件。

在 `envs/` 目录下，复制或编辑对应环境的 `.env` 文件（例如 `envs/csu.env`）。您**必须**修改 `MASTER_IP` 和 `HARBOR` 这两个关键变量：

    _env_name: "{{ env_name | default('p368') }}"

    _deploy_dir: "{{ deploy_dir | default('/opt/csst-airflow') }}"

  tasks:

    - name: Ensure envs directory exists

      file:

        path: "{{ _deploy_dir }}/envs"

        state: directory

    - name: Render .env file from template

      template:

        src: "{{ playbook_dir }}/roles/common/templates/env.j2"

        dest: "{{ _deploy_dir }}/envs/{{ _env_name }}.env"

    - name: Start Master services (postgres, redis, elasticsearch, airflow core, api, etc.)

      shell: |

        {{ docker_compose_cmd }} --env-file envs/{{ _env_name }}.env --profile master up -d

    _env_name: "{{ env_name | default('p368') }}"

    _deploy_dir: "{{ deploy_dir | default('/opt/csst-airflow') }}"

  tasks:

    - name: Ensure envs directory exists

      file:

        path: "{{ _deploy_dir }}/envs"

        state: directory

    - name: Render .env file from template

      template:

        src: "{{ playbook_dir }}/roles/common/templates/env.j2"

        dest: "{{ _deploy_dir }}/envs/{{ _env_name }}.env"

    - name: Start Worker services (celery-worker, filebeat, monitoring)

      shell: |

        export AIRFLOW__CELERY__WORKER_CONCURRENCY={{ worker_concurrency | default(16) }}

[master]

# Replace with your actual master node IP

# Replace with your actual master node IP or hostname

csu-master ansible_host=192.168.25.18 ansible_user=root

[worker]

# Add all your worker node IPs here

# e.g., csu-worker1 ansible_host=192.168.25.19 ansible_user=root

# Add one worker node per line

# e.g., csu-worker1 ansible_host=192.168.25.19 ansible_user=root deploy_dir=/opt/csst-airflow-worker1 worker_concurrency=16

# e.g., csu-worker2 ansible_host=192.168.25.20 ansible_user=root deploy_dir=/opt/csst-airflow-worker2 worker_concurrency=32

[all:vars]

# The environment name to deploy (e.g., p368, csu, zjlab)

[master]

# Replace with your actual master node IP

# Replace with your actual master node IP or hostname

zjlab-master ansible_host=10.200.60.244 ansible_user=root

[worker]

# Add all your worker node IPs here

# e.g., zjlab-worker1 ansible_host=10.200.60.245 ansible_user=root

# Add one worker node per line

# e.g., zjlab-worker1 ansible_host=10.200.60.245 ansible_user=root deploy_dir=/opt/csst-airflow-worker1 worker_concurrency=16

# e.g., zjlab-worker2 ansible_host=10.200.60.246 ansible_user=root deploy_dir=/opt/csst-airflow-worker2 worker_concurrency=32

[all:vars]

# The environment name to deploy (e.g., p368, csu, zjlab)

# --- Celery Worker Config ---

# Defines the concurrency per worker node. Can be overridden in inventory.

WORKER_CONCURRENCY={{ worker_concurrency | default(8) }}

WORKER_LABEL={{ ansible_host }}