rebrand: kusama → dicle

- Replace all kusama/Kusama references with dicle/Dicle
- Rename weight files from ksm_size to dcl_size
- Update papi-tests files from ksm to dcl
- Remove chain-specs/kusama.json files
- cargo check --workspace successful (Finished output)
- Update MAINNET_ROADMAP.md: FAZ 8 completed

pezcumulus/teyrchains/runtimes/bridge-hubs/bridge-hub-pezkuwichain/src/bridge_common_config.rs

@@ -123,9 +123,9 @@ impl pezpallet_bridge_grandpa::Config<BridgeGrandpaPezkuwichainBulletinInstance>
 	type HeadersToKeep = RelayChainHeadersToKeep;
 	// Technically this is incorrect - we have two pezpallet instances and ideally we shall
 	// benchmark every instance separately. But the benchmarking engine has a flaw - it
-	// messes with components. E.g. in Kusama maximal validators count is 1024 and in
+	// messes with components. E.g. in Dicle maximal validators count is 1024 and in
 	// Bulletin chain it is 100. But benchmarking engine runs Bulletin benchmarks using
-	// components range, computed for Kusama => it causes an error.
+	// components range, computed for Dicle => it causes an error.
 	//
 	// In practice, however, GRANDPA pezpallet works the same way for all bridged chains, so
 	// weights are also the same for both bridges.