[danoswaltCL]

• adds alpha param to ts configurable
• db migration for the config table
• adds scaling calculation and tooltips to feedback area

[Copilot]

🟡 Changes recommended

Configuration edits can write null to a non-nullable scaling-factor column and leave the feedback card displaying stale posteriors and estimated weights.

2 open issues

• Normalize null scalingFactor before updating the non-nullable column · New
• Refresh rewards summary after configuration edits · New

What changed in this PR

Adds a configurable scaling factor (α) to UpGrade’s native Thompson Sampling implementation. It adjusts the influence of reward counts without changing priors, while keeping configuration and feedback displays aligned with the calculation.

Changes:

• Persists α with a default of 1 and applies it to assignment and reward-summary posteriors.
• Adds form validation, overview values, scaled counts, and explanatory tooltips.
• Extends shared types, unit tests, and implementation documentation.

File	Description
interfaces.ts	Adds scaling factor and posterior response fields.
en.json	Adds labels, validation messages, and formula tooltips.
common-form-helpers.service.ts	Adds decimal-place validation.
ts-configurable-reward-count-table.component.ts	Formats scaled counts and posterior values.
ts-configurable-reward-count-table.component.scss	Adjusts count spacing and wrapping.
ts-configurable-reward-count-table.component.html	Displays scaled counts, posteriors, and tooltips.
experiment-reward-feedback-section-card.component.html	Displays α and passes it to the table.
ts-configurable-policy-parameters-form.component.ts	Adds the scaling-factor form control.
ts-configurable-policy-parameters-form.component.html	Adds the input and validation messages.
thompson-sampling-helper.service.ts	Handles defaults, validation, configuration, and overview formatting.
thompson-sampling-helper.service.spec.ts	Tests scaling-factor helpers and validation.
experiments.selectors.ts	Includes α in fallback reward summaries.
experiments.selector.spec.ts	Updates summary expectations.
experiments.reducer.spec.ts	Updates reward-summary fixtures.
experiments.model.ts	Extends configuration types and overview labels.
experiments.effects.spec.ts	Updates the reward-fetch fixture.
ThompsonSamplingService.test.ts	Tests scaled sampling and unchanged priors.
ThompsonSamplingExperimentCrudService.test.ts	Tests persistence, defaults, and scaled summaries.
ExperimentAssignmentService.test.ts	Tests passing α to the algorithm.
ExperimentDTO.test.ts	Tests scaling-factor input constraints.
1794510446000-thompsonSamplingScalingFactor.ts	Adds the database column with default 1.
ThompsonSamplingService.ts	Applies α to reward counts in posterior calculations.
ThompsonSamplingExperimentCrudService.ts	Persists α and returns scaled posterior summaries.
ExperimentAssignmentService.ts	Passes configured α into sampling.
ThompsonSamplingExperimentConfig.ts	Defines the scaling-factor entity column.
ExperimentDTO.ts	Adds numeric bounds and precision validation.
CLAUDE.md	Documents scaling semantics and UI behavior.

_{Review effort: Balanced}