Allure Dashboard overview shows incorrect testcase count with Pabot Robot Framework runs

### Describe the Bug

When running Robot Framework tests for multiple websites in parallel using Pabot (with argument files), the Allure Dashboard overview displays only the testcase count for a single URL instead of the total count across all URLs. For example, with 11 testcases per URL and 2 URLs, the timeline correctly shows 22 testcases, but the overview only shows 11.

### Steps to Reproduce

1. Run Robot Framework tests in parallel using Pabot with argument files for multiple URLs:
pabot --listener "allure_robotframework;allure-results" --argumentfile1 arg1.txt --argumentfile2 arg2.txt tests/Global
2. Open the generated Allure report.
3. Observe the timeline tab: it shows the correct total testcase count (e.g., 22).
4. Observe the overview tab: it only shows the testcase count for a single URL (e.g., 11), not the total.

### Expected Behaviour

- The Dashboard overview should display the total number of executed testcases from all parallel runs (i.e., sum across all URLs/sites), matching the count shown in the timeline.
- The pass/fail count should display the total number (sum of pass/fail across all URLs).

### Screenshots or Additional Context

Allure Dashboard - Overview Tab showing only 11 testcase of single url

Allure Dashboard - Timeline Tab showing total 22 testcase of two urls

Environment:
- Allure Commandline version: 2.34.1
- Allure RobotFramework listener version: 2.15.0
- Robot Framework version: 6.1.1
- Pabot version: 4.1.1
- OS: Windows
- Browser: Chrome
- Java version: 24.0.2

### What Language are you using?

Python (RobotFramework)

### What Framework/Allure Integration you are using?

allure-robotframework

### What version of Allure Integration you are using?

2.15.0

### What version of Allure Report you are using?

2.34.1

### Code of Conduct

- [x] I agree to follow this project's Code of Conduct

Commencez par reproduire la commande Pabot signalée avec les deux fichiers d’arguments et examinez les données overview et timeline du rapport Allure généré. Suivez la façon dont les résultats allure-robotframework des exécutions parallèles sont agrégés ; le travail est terminé lorsque les totaux de cas de test et de pass/fail de overview correspondent aux totaux de timeline pour toutes les URLs.

Rédigé par le modèle d'indexation à partir du texte de l'issue.