Add gazetteer configuration and dependency

This commit introduces the `VOCAB_DIR` configuration option, which
specifies the directory for gazetteer files. It also adds the `strsim`
crate, which is used for calculating string similarity and is essential
for the gazetteer's functionality.

The gazetteer module has been significantly expanded to include the
`Gazetteer` struct, its loading mechanism from `.txt` files, and an
annotation function. This functionality allows for pre-LLM correction of
proper names and technical vocabulary by identifying potential
misspellings based on phonetic similarity and edit distance.

server/src/config.rs

@@ -58,6 +58,10 @@ pub struct Config {
     /// or empty. Runtime-configurable so admins can iterate and use the
     /// "Neu analysieren" button to re-run cases.
     pub llm_system_prompt: String,
+    /// Directory containing gazetteer `*.txt` files (anatomy, substances,
+    /// medications). Loaded once at startup; a missing directory is a
+    /// non-fatal WARN — analysis runs without proper-name correction.
+    pub vocab_dir: PathBuf,
     pub session_timeout_hours: u32,
     /// Whether to set the `Secure` flag on the session cookie. Default `true`.
     /// Set `COOKIE_SECURE=false` only for plain-HTTP local development —
@@ -100,6 +104,7 @@ impl Config {
                     v
                 }
             },
+            vocab_dir: PathBuf::from(optional_env("VOCAB_DIR", "./vocab")),
             session_timeout_hours: optional_env_parsed("SESSION_TIMEOUT_HOURS", 8),
             cookie_secure: optional_env_parsed("COOKIE_SECURE", true),
         }
@@ -152,6 +157,7 @@ impl Config {
             llm_temperature: 0.0,
             llm_timeout_seconds: 180,
             llm_system_prompt: crate::analyze::prompt::SYSTEM_PROMPT.to_string(),
+            vocab_dir: PathBuf::new(),
             session_timeout_hours: 8,
             cookie_secure: false,
         }