Expose MinimumPeakHeight to CalibrateRectangularDetectors

[Peter Peterson]

This is a good way to filter out bad peaks that is already in the code, just not easily accessible to the user.

Also be more consistent on how things are returned from FindPeaks and GetDetOffsetMultiPeaks, how backgrounds are calculated, the d-range of interest for fitting, and clarify the FitInfo table (aka, little bit of code cleanup).

[Peter Peterson]

Re #7736. Consistently use the same code for bad fits.

Changeset: 8abc14005a282dbf5ce6bb949683d06babe0fd24

[Peter Peterson]

Re #7736. Actually use the MinimumPeakHeight parameter.

It was passed into FindPeaks, but not checked again in GetDetOffsetsMultiPeaks. Also check for negative chisq being being returned.

Changeset: 851077b9ac5253b2f619fd142b9b230b48672512

[Peter Peterson]

Re #7736. Exposing MinimumPeakHeight through py-algorithm.

Changeset: 6bf96aba53d9b4403a9ca077aeec971d30ba9e4b

[Peter Peterson]

Re #7736. Refactored FindPeaks to simplify code (slightly).

This included:

1. Only calculate the peak center once
2. If the guessed peak width is < 1e-200 then don't bother fitting
3. estimatePeakParameters just returns the error message if there is one.

Changeset: f9a03e37ddcfb2645c78637e71d4e5039ad45d3c

[Peter Peterson]

Re #7736. Reduce the d-range that is used for finding peaks.

This is really just looking for where there is actual data (non-zero counts) and fitting that. There are also changes for reducing the number of uninteresting log messages.

Changeset: 14ba62cb9e107567dd8ecf760907dd34065ea63a

[Peter Peterson]

Re #7736. Cleaned up information in InfoTableWorkspace

Changeset: 7b3f83aaa91e59d4f8ec1fb0c0d71193aca5034a

[Peter Peterson]

Re #7736. Bringing all background estimations into one place

Changeset: 23312e4caefeed3686968e66ac4d4330e8f321e6

[Peter Peterson]

This is on ​feature/7736_calibrate_pg3.

To test: The main feature to test is that you can still calibrate the instrument without specifying MinimumPeakHeight and that the FitInfoTable shows a lot less fitted peaks when you increase the number.

[Peter Peterson]

Re #7736. Updating golden results for PG3 calibration.

Changeset: e470736493e73ed33107e78ec9592067eab34bc8

[Peter Peterson]

This is on ​mantid:feature/7736_calibrate_pg3 and ​systemtests:feature/7736_calibrate_pg3.

To test: The main feature to test is that you can still calibrate the instrument without specifying MinimumPeakHeight and that the FitInfoTable shows a lot less fitted peaks when you increase the number.

[Wenduo Zhou]

By running the system test CalibrateRectangularDetector_Test with different MinimumPeakHeight, the number of peaks masked increases with higher minimum peak height. Ticket is thus close.

[Wenduo Zhou]

Merge remote-tracking branch 'origin/feature/7736_calibrate_pg3'

[Martyn Gigg]

Merge remote-tracking branch 'origin/feature/7736_calibrate_pg3'

[Stuart Campbell]

This ticket has been transferred to github issue ​8581