Allen RICH decoding

This MR introduces RICH decoding, concretely producing RichSmartIDs from RICH1 and RICH2 raw banks.

• Created algorithm rich_decoding_t, which performs the decoding by calculating the number of hits, prefix sum and finally decoding the hits.
• Added simplified copies of RichSmartID, PackedFrameSizes, Tel40CableMapping and PDMDBDecodeMapping adjusted to the requirements of GPU computing, namely no dynamic memory allocation, no std::string elements and __host__ __device__ attributes.
• Created Allen::RawEvent and Allen::RawBank which replace CaloRawEvent and CaloRawBank. Ideally this should be used in other subdetectors as well due to the raw bank format being mostly identical.
• Added configurations rich and hlt1_pp_rich_no_ut.
• Added throughput test of hlt1_pp_rich_no_ut.
• Added two geometry files corresponding to Tel40CableMapping and PDMDBDecodeMapping, alongside constants and the necessary scaffolding.

Goes with LHCb!4001 (merged) and Moore!2171 (merged)

Done in collaboration with @caschbur .

Dumpers/BinaryDumpers/CMakeLists.txt

@@ -21,7 +21,8 @@ gaudi_add_library(BinaryDumpers
                     LHCb::MagnetLib
                     LHCb::FTDetLib
                     LHCb::RecEvent
-                    LHCb::MuonDAQLib)
+                    LHCb::MuonDAQLib
+                    LHCb::RichFutureDAQLib)
 
 gaudi_add_module(BinaryDumpersModule
                  SOURCES