execute(value='response') and why_not(..., value='response') refuse the documented value #142

What happens

value='response' is in the signature and docstring of Study.execute (value: Literal['response', 'R1', …]; "None or 'response': Chart the response variable (default)"), but every chart refuses it:

study.execute(chart='Xbar', value='response')
# ChartNotAvailableError: 'Xbar' with value='response' is not valid for ADS 1.
# Valid charts for RESPONSE: None

study.why_not('Xbar', value='response')
# "'response' is not a recognized residual. Valid values: R1, R2, R3, R4, R5, R6 …"

value=None works, so callers that leave value out never see it.

Why

_validate_execute_request (processbehavior/study.py, ~2279) decides whether a value is a residual with value.upper().startswith('R'), which is true for 'RESPONSE'. The value then goes down the residual path and fails the residual pairing check. why_not and supports_calibration take the same value and refuse it the same way.

Proposed change

Normalise 'response' (any case) to None once, at the start of execute(), supports_calibration() and why_not(). _resolve_value_column already treats None as the response. Tests: value='response', 'Response' and None give identical results on Xbar, S, X and Histogram, and why_not and supports_calibration agree.

Context

Found while scoping phased limits for Xbar/S.