Fail asv workflows when performance deteriorates

Nadie ha tomado este issue todavía.

Fail the ASV workflows if at least one of benchmarks deteriorates above a certain threshold (e.g. 150% / 200%).

**asv-main.yml**:
- Execute `asv run` without `|| true`, allowing the command to exit with a non-zero return code
- Submit benchmarks and deploy dashboard on gh-pages, if the previous command succeeded or not

**asv-pr.yml**:
- Execute `asv continuous` without `|| true`, allowing the command to exit with a non-zero return code
- Output the performance comment on the PR if the comparison was successful or not

**asv-nightly.yml**
- Execute `asv continuous` without `|| true`, allowing the command to exit with a non-zero return code
- Write the commit hash to cache, if the previous if the comparison was successful or not

More information about "on failure" executions [here](https://docs.github.com/en/actions/learn-github-actions/expressions#status-check-functions).

Comience con .github/workflows/asv-main.yml, asv-pr.yml y asv-nightly.yml, e inspeccione cómo gestiona actualmente cada uno los errores de asv run o asv continuous. Actualice las condiciones de error de los workflows preservando los pasos solicitados de submit, dashboard, PR comment y cache; después, verifique que los workflows sigan ejecutando esos pasos cuando fallen las comparaciones.

Escrito por el modelo de indexación a partir del texto del issue.