⚡ Improve handling of dedicated subgraphs in exact mapper (#308)

## Description

This tiny PR improves the performance when a dedicated subgraph of an
architecture is specified. This was only incompletely handled, which led
to very inefficient computations (e.g., the SWAP limit was still
computed from the whole architecture).
This PR just reduces the architecture to the subgraph as part of the
mapping.

## Checklist:

<!---
This checklist serves as a reminder of a couple of things that ensure
your pull request will be merged swiftly.
-->

- [x] The pull request only contains commits that are related to it.
- [x] I have added appropriate tests and documentation.
- [x] I have made sure that all CI jobs on GitHub pass.
- [x] The pull request introduces no new warnings and follows the
project's style guidelines.

Signed-off-by: Lukas Burgholzer <[email protected]>

src/exact/ExactMapper.cpp

@@ -56,9 +56,7 @@ void ExactMapper::map(const Configuration& settings) {
     return;
   }
 
-  // 2) For all possibilities k (=m over n) to pick n qubits from m physical
-  // qubits
-  std::vector<std::uint16_t> qubitRange{};
+  // 2a) If only a subgraph should be considered, reduce the architecture.
   if (!config.subgraph.empty()) {
     const auto subgraphQubits = config.subgraph.size();
     if (subgraphQubits < qc.getNqubits()) {
@@ -76,20 +74,23 @@ void ExactMapper::map(const Configuration& settings) {
       std::cerr << "The subgraph is not connected." << std::endl;
       return;
     }
-    qubitRange = Architecture::getQubitList(reducedCouplingMap);
-  } else {
-    qubitRange.clear();
-    auto qubitSet = architecture.getQubitSet();
-    qubitRange.reserve(qubitSet.size());
-    std::copy(qubitSet.begin(), qubitSet.end(), std::back_inserter(qubitRange));
+
+    architecture.setCouplingMap(reducedCouplingMap);
   }
+
+  // 2b) If configured to use subsets, collect all k (=m over n) possibilities
+  // to pick n qubits from m device qubits. Otherwise, consider all qubits.
+  std::vector<std::uint16_t> qubitRange =
+      Architecture::getQubitList(architecture.getCouplingMap());
+
   std::vector<QubitChoice> allPossibleQubitChoices{};
   if (config.useSubsets) {
     allPossibleQubitChoices = architecture.getAllConnectedSubsets(
         static_cast<std::uint16_t>(qc.getNqubits()));
   } else {
     allPossibleQubitChoices.emplace_back(qubitRange.begin(), qubitRange.end());
   }
+
   // 3) determine exact mapping for this qubit choice
   std::vector<Swaps> swaps(reducedLayerIndices.size(), Swaps{});
   mappingSwaps.reserve(reducedLayerIndices.size());