logic: add forecast grid charge target strategy

[filiplajszczak]

Adds an opt-in forecast grid-charge target strategy while keeping the default fixed behavior unchanged.

With forecast, min_grid_charge_soc remains a floor, and batcontrol raises the effective target from forecasted expensive-slot net demand, capped by max_charging_from_grid_limit.

Also publishes the runtime effective target via MQTT/Home Assistant discovery.

I’ll add the corresponding wiki documentation after merge.

[Copilot]

Pull request overview

This PR adds an opt-in forecast strategy for determining the effective minimum grid-charge SoC target at runtime, while keeping the default fixed behavior unchanged. The effective target is computed from forecasted net demand during upcoming expensive price slots (with optional PV forecast discounting) and is then passed into the logic layer and exposed via MQTT.

Changes:

• Introduces batcontrol.logic.grid_charge_target.calculate_effective_grid_charge_soc() with fixed and forecast strategies.
• Adds core config parsing and per-run computation of effective_min_grid_charge_soc, passing it into CalculationParameters and publishing it via MQTT.
• Extends MQTT API + discovery and adds unit/integration-style tests for the new behavior.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
src/batcontrol/logic/grid_charge_target.py	New module implementing effective grid-charge SoC target calculation strategies.
src/batcontrol/core.py	Parses new config, computes effective target each run, publishes it to MQTT, and feeds it into logic parameters.
src/batcontrol/mqtt_api.py	Publishes the effective target and exposes it via Home Assistant MQTT discovery.
tests/batcontrol/logic/test_grid_charge_target.py	Unit tests for fixed/forecast target computation and capping behavior.
tests/batcontrol/test_core.py	Validates config parsing and verifies core passes/publishes the effective target.
tests/batcontrol/test_mqtt_api.py	Tests MQTT publishing + discovery for the new effective target sensor.
config/batcontrol_config_dummy.yaml	Documents the new optional config keys in the dummy config.

[filiplajszczak]

Addressed in f5261de: sparse/non-consecutive forecast dict indices now raise a clear ValueError, covered by a regression test.

[filiplajszczak]

Addressed the two Copilot comments in f5261de: forecast dict indices now fail with a clear ValueError/regression test, and grid_charge_target_strategy parsing now strips whitespace before lowercasing.

[MaStr]

I tried to move all these calculations over to the logic/ packages and have core.py more like a wiring and function router place.

With this context, does it make sense to have it here in the code, or not?

[filiplajszczak]

Good point. Addressed in 690a32a: the grid-charge target config parsing and effective-target resolver now live in batcontrol.logic.grid_charge_target; core.py only builds CalculationInput, calls the logic helper, publishes the effective target, and passes it into CalculationParameters.

[MaStr]

Ok, I did another review and try to understand what you did there.
in the default logic, there is the recharge amount defined by the high price slot.
Why not just add the capacity on top of the recharge amount during the recharge energy calculation?