Adjusts the volume control for each speaker in the mixer platform .

**Describe the problem you have/What new integration you would like**

When testing voice satellites on esp32s3 I encounter cases where the volume of tts engines can differ from the average music level, which is not very practical. Fine tuning for each _source_speaker_ would be very helpful for such cases.

**Please describe your use case for this integration and alternatives you've tried:**

I know that we can currently influence this parameter through the _mixer_speaker.apply_ducking_ action, but it would be preferable to specify a value for the channel in the configuration file to precisely align the volumes. Or additionally, adding an action to control this base value would allow it to be managed from HA.

**Additional context**

As far as I understand, the feature itself is already present under the hood of esphome, the only thing missing is the implementation of controls for the end user.

Thanks for your work.

No source files or tests are named. Start by tracing the existing mixer_speaker.apply_ducking action and the ESPHome configuration and action interfaces, then verify the claim that per-source volume behavior already exists. Done means users can set and manage a distinct base volume for each source speaker.

Written by the indexing model from the issue text.