Enterprise PR: https://github.com/odoo/enterprise/pull/84122

Since commit ae16d9d, the Adam driver priority has been incorrectly set higher than it should be. The issue was a side-effect of removing the meta-class, since it no longer registered the `Driver` base class as a driver.

The fix was just to simplify the priority system to not be affected by inheritance. Then we manually set the Adam priority to -1, and the Blackbox priority to 1. Everything else will have priority 0.


For a detailed breakdown of the old behaviour and the cause of the bug, keep reading:

The priority system was influenced by inheritance - the more levels of inheritance, the higher the priority. For example, in saas-18.1, you get these priorities:
- `SerialBaseDriver`: 1
- `BlackBoxDriver`: 2
- `ScaleDriver`: 2
- `Toledo8217Driver`: 3

The Adam driver sets its priority to zero, but it still gets +1 from the meta class. So the final priorities are as follows in saas-18.1:
- `SerialBaseDriver`: 1
- `AdamEquipmentDriver`: 1
- `BlackBoxDriver`: 2
- `ScaleDriver`: 2
- `Toledo8217Driver`: 3

This results in the Adam driver runs last.

In saas-18.2, the priorities are shifted by 1 due to the `Driver` base class not being registered, but the Adam driver still ends up with the same value. So the priorities now look like this:
- `SerialBaseDriver`: 0
- `AdamEquipmentDriver`: 1
- `BlackBoxDriver`: 1
- `ScaleDriver`: 1
- `Toledo8217Driver`: 2

The Adam driver and blackbox driver have the same priority, so the Adam driver could end up running first and breaking the blackbox.

task-4750364

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr

SQL traceback occurs when user creates their own automation rule involving python code.

**Steps to reproduce:**
Install base_automation,discuss if not already installed
* `Setting> Technical>Automation rules`
* name it as `Whatsapp add admin`
* model as `discuss channel` and trigger `after creation` any amount of time
* `Add an action>Execute code`
* Add the following code:  

```python
try: 
 env['discuss.channel.member'].create({ 'partner_id': 1,  # Likely invalid 'channel_id': 1, }) 
except Exception: 
 pass

env['res.partner'].search([], limit=1)
```

* Go to `techinical>automation>scheduled actions>Automation rules:check and execute` run it manually.

`ValueError:InFailedSqlTransaction('current transaction is aborted, commands ignored until end of transaction block\n') while evaluating
'model._cron_process_time_based_actions()'`

**Solution:**
* Place try and except inside the loop to handle errors per record and allow the loop to continue and handle the transaction error better.
* This will only one occurence of the error instead of repeating occurrence each time the cron runs.

**Sentry-6562754804**

---
I confirm I have signed the CLA and read the PR guidelines at www.odoo.com/submit-pr