perf: eliminate ~70 unnecessary sqrt ops per frame, optimize caches and threading

Squared distance optimizations across 30 files:
- Convert glm::length() comparisons to glm::dot() (no sqrt)
- Use glm::inversesqrt() for check-then-normalize patterns (1 rsqrt vs 2 sqrt)
- Defer sqrt to after early-out checks in collision/movement code
- Hottest paths: camera_controller (21), weather particles, WMO collision,
  transport movement, creature interpolation, nameplate culling

Container and algorithm improvements:
- std::map<string> → std::unordered_map for asset/DBC/MPQ/warden caches
- std::mutex → std::shared_mutex for asset_manager and mpq_manager caches
- std::sort → std::partial_sort in lighting_manager (top-2 of N volumes)
- Double-lookup find()+operator[] → insert_or_assign in game_handler
- Add reserve() for per-frame vectors: weather, swim_effects, WMO/M2 collision

Threading and synchronization:
- Replace 1ms busy-wait polling with condition_variable in character_renderer
- Move timestamp capture before mutex in logger
- Use memory_order_acquire/release for normal map completion signaling

API additions:
- DBC getStringView()/getStringViewByOffset() for zero-copy string access
- Parse creature display IDs from SMSG_CREATURE_QUERY_SINGLE_RESPONSE

include/game/warden_emulator.hpp

@@ -5,6 +5,7 @@
 #include <string>
 #include <memory>
 #include <map>
+#include <unordered_map>
 #include <functional>
 
 // Forward declare unicorn types (will include in .cpp)
@@ -148,7 +149,7 @@ private:
     uint32_t apiStubBase_;           // API stub base address
 
     // API hooks: DLL name -> Function name -> stub address
-    std::map<std::string, std::map<std::string, uint32_t>> apiAddresses_;
+    std::unordered_map<std::string, std::unordered_map<std::string, uint32_t>> apiAddresses_;
 
     // API stub dispatch: stub address -> {argCount, handler}
     struct ApiHookEntry {

include/game/world_packets.hpp

@@ -1502,9 +1502,10 @@ struct CreatureQueryResponseData {
     std::string subName;
     std::string iconName;
     uint32_t typeFlags = 0;
-    uint32_t creatureType = 0;
+    uint32_t creatureType = 0;  // 1=Beast, 2=Dragonkin, 3=Demon, 4=Elemental, 5=Giant, 6=Undead, 7=Humanoid, ...
     uint32_t family = 0;
     uint32_t rank = 0;         // 0=Normal, 1=Elite, 2=Rare Elite, 3=Boss, 4=Rare
+    uint32_t displayId[4] = {};  // Up to 4 random display models (0 = unused)
 
     bool isValid() const { return entry != 0 && !name.empty(); }
 };

include/pipeline/asset_manager.hpp

@@ -8,7 +8,9 @@
 #include <string>
 #include <vector>
 #include <map>
+#include <unordered_map>
 #include <mutex>
+#include <shared_mutex>
 
 namespace wowee {
 namespace pipeline {
@@ -164,15 +166,15 @@ private:
      */
     std::string resolveFile(const std::string& normalizedPath) const;
 
-    mutable std::mutex cacheMutex;
-    std::map<std::string, std::shared_ptr<DBCFile>> dbcCache;
+    mutable std::shared_mutex cacheMutex;
+    std::unordered_map<std::string, std::shared_ptr<DBCFile>> dbcCache;
 
     // File cache (LRU, dynamic budget based on system RAM)
     struct CachedFile {
         std::vector<uint8_t> data;
         uint64_t lastAccessTime;
     };
-    mutable std::map<std::string, CachedFile> fileCache;
+    mutable std::unordered_map<std::string, CachedFile> fileCache;
     mutable size_t fileCacheTotalBytes = 0;
     mutable uint64_t fileCacheAccessCounter = 0;
     mutable size_t fileCacheHits = 0;

include/pipeline/dbc_loader.hpp

@@ -3,6 +3,7 @@
 #include <vector>
 #include <map>
 #include <string>
+#include <string_view>
 #include <cstdint>
 #include <memory>
 
@@ -92,6 +93,11 @@ public:
      */
     std::string getString(uint32_t recordIndex, uint32_t fieldIndex) const;
 
+    /**
+     * Get a string field as a view (no allocation; valid while this DBCFile lives)
+     */
+    std::string_view getStringView(uint32_t recordIndex, uint32_t fieldIndex) const;
+
     /**
      * Get string by offset in string block
      * @param offset Offset into string block
@@ -99,6 +105,11 @@ public:
      */
     std::string getStringByOffset(uint32_t offset) const;
 
+    /**
+     * Get string by offset as a view (no allocation; valid while this DBCFile lives)
+     */
+    std::string_view getStringViewByOffset(uint32_t offset) const;
+
     /**
      * Find a record by ID (assumes first field is ID)
      * @param id Record ID to find

include/pipeline/mpq_manager.hpp

@@ -8,6 +8,7 @@
 #include <unordered_map>
 #include <unordered_set>
 #include <mutex>
+#include <shared_mutex>
 
 // Forward declare StormLib handle
 typedef void* HANDLE;
@@ -115,7 +116,7 @@ private:
     //
     // Important: caching misses can blow up memory if the game probes many unique non-existent filenames.
     // Miss caching is disabled by default and must be explicitly enabled.
-    mutable std::mutex fileArchiveCacheMutex_;
+    mutable std::shared_mutex fileArchiveCacheMutex_;
     mutable std::unordered_map<std::string, HANDLE> fileArchiveCache_;
     size_t fileArchiveCacheMaxEntries_ = 500000;
     bool fileArchiveCacheMisses_ = false;

include/rendering/character_renderer.hpp

@@ -13,6 +13,7 @@
 #include <utility>
 #include <future>
 #include <deque>
+#include <condition_variable>
 #include <mutex>
 #include <atomic>
 
@@ -325,6 +326,7 @@ private:
     };
     // Completed results ready for GPU upload (populated by background threads)
     std::mutex normalMapResultsMutex_;
+    std::condition_variable normalMapDoneCV_;  // signaled when pendingNormalMapCount_ reaches 0
     std::deque<NormalMapResult> completedNormalMaps_;
     std::atomic<int> pendingNormalMapCount_{0};  // in-flight background tasks