Merge bitcoin/bitcoin#30980: fuzz: fix bug in p2p_headers_presync harness

a7498cc7e2 Fix bug in p2p_headers_presync harness (marcofleon)

Pull request description:

  The calculation for the test chain's work (`total_work`) should be outside of the loop. Previously, `total_work` was being miscalculated due to multiple additions of work from the same headers. Now, each header's work is only counted once, providing an accurate total.

  https://github.com/bitcoin/bitcoin/pull/30918 followup

ACKs for top commit:
  dergoegge:
    utACK a7498cc7e2
  instagibbs:
    ACK a7498cc7e2
  glozow:
    makes sense, utACK a7498cc7e2
  mzumsande:
    ACK a7498cc7e2

Tree-SHA512: b95f25dcf7ace220e30f1d72f50d85ee18777467927c0cc1ed8582b390cb7185ffc0e2f127309eb083044fb41f5a13fce5ebb15b7952718a899bafff26921be8

src/test/fuzz/p2p_headers_presync.cpp

@@ -197,6 +197,7 @@ FUZZ_TARGET(p2p_headers_presync, .init = initialize)
                 auto headers_msg = NetMsg::Make(NetMsgType::BLOCK, TX_WITH_WITNESS(block));
                 g_testing_setup->SendMessage(fuzzed_data_provider, std::move(headers_msg));
             });
+    }
 
     // This is a conservative overestimate, as base is only moved forward when sending headers. In theory,
     // the longest chain generated by this test is 1600 (FUZZ_MAX_HEADERS_RESULTS * 100) headers. In that case,
@@ -205,7 +206,6 @@ FUZZ_TARGET(p2p_headers_presync, .init = initialize)
 
     // This test should never create a chain with more work than MinimumChainWork.
     assert(total_work < chainman.MinimumChainWork());
-    }
 
     // The headers/blocks sent in this test should never be stored, as the chains don't have the work required
     // to meet the anti-DoS work threshold. So, if at any point the block index grew in size, then there's a bug