Correct some inconsistencies in V16 geometry

[EmyrClement]

PR description:

This PR fixes two inconsistencies with the V16 trigger geometry:

• The wafer centring of the CE-H were assumed to all be corner centred, but some are not wafer centred. These cases are checked against the information from the topology class.
• HGCalTriggerModuleDetId only distinguishes between fine and coarse scintillator modules, however HGCScintillatorDetId also distinguishes between type "c" and "m" of coarse modules. For now, I removed this distinction.

PR validation:

Checked the number of errorbits reported by the HGCalTriggerGeoTesterBackendStage1 tester has reduced. There are still inconsistencies, but fewer than before.

[jbsauvan]

I'm wondering, since there is potentially some other changes between $<V16$ and $\geq V16$ geometries (potentially incompatible changes?), if it would make sense to write a new geometry class, to be used for post-V16 geometries ([...]GeometryV16[...])?

[jbsauvan]

Do we still need to separate between odd and even layers if we can retrieve the type of layer centring with dddConstants.layerType()?

[EmyrClement]

Looking at where dddConstants.layerType() retrieves the layer type:

cmssw/Geometry/HGCalCommonData/src/HGCalTypes.cc

Lines 42 to 48 in 7724d3d

	int32_t HGCalTypes::layerType(int type) {
	static constexpr int32_t types[5] = {HGCalTypes::WaferCenter,
	HGCalTypes::WaferCenterB,
	HGCalTypes::CornerCenterYp,
	HGCalTypes::CornerCenterYm,
	HGCalTypes::WaferCenterR};
	return ((type >= 0 && type < 5) ? types[type] : HGCalTypes::WaferCenter);

If HGCalTypes::CornerCenterYp and HGCalTypes::CornerCenterYm uniquely correspond to one of HGCalGeomRotation::WaferCentring::CornerCentredY and HGCalGeomRotation::WaferCentring::CornerCentredMercedes, then I think we can remove the separation between odd and even layers. Though I don't know which of CornerCenterYp/m corresponds to CornerCentredY/Mercedes - I can work this out, unless you happen to know.

[jbsauvan]

I would naively guess that Yp is Y and Ym is Mercedes (negative version of Y), but not sure at all.

[EmyrClement]

@jbsauvan I see that the jenkins tests have failed when running the cmssw config with these changes. However I can't access the link to get more information to understand why/see a more detailed log - are you able to see what happened?

[jbsauvan]

Hi @EmyrClement
I checked on Jenkins but cannot find more info. Seems like something related to the validation code and unrelated to this PR. I'm checking with the validation developer.

@jbsauvan I see that the jenkins tests have failed when running the cmssw config with these changes. However I can't access the link to get more information to understand why/see a more detailed log - are you able to see what happened?

[EmyrClement]

Thanks - if it's of any use, I see the same error with #113

Hi @EmyrClement I checked on Jenkins but cannot find more info. Seems like something related to the validation code and unrelated to this PR. I'm checking with the validation developer.

@jbsauvan I see that the jenkins tests have failed when running the cmssw config with these changes. However I can't access the link to get more information to understand why/see a more detailed log - are you able to see what happened?