[SPARK-57756][SQL] Move SQL UDF resolution SQLConf construction to common code

SessionCatalog.makeSQLFunctionPlan and the ResolveSQLFunctions / ResolveSQLTableFunctions analyzer rules each inline the same throwaway-SQLConf construction used to resolve a SQL UDF body: new SQLConf, seed from the function's stored configs, then either set ANSI unconditionally or overlay the session's retained resolution configs. Move it into a common Analyzer.buildSQLFunctionConf helper, parameterized by alwaysSetAnsiValue and applySessionOverrides, and have the three call sites delegate to it. Behavior-preserving.

### What changes were proposed in this pull request?

`SessionCatalog.makeSQLFunctionPlan` and the `ResolveSQLFunctions` /
`ResolveSQLTableFunctions` analyzer rules each inline the same throwaway-`SQLConf`
construction used to resolve a SQL UDF body. This moves it into a common
`Analyzer.buildSQLFunctionConf` helper, parameterized by `alwaysSetAnsiValue` and
`applySessionOverrides`, and has the three call sites delegate to it.

### Why are the changes needed?

The three call sites carried the same conf-seeding logic. Unifying it removes the
duplication and gives one place to reason about a SQL UDF body's resolution conf.

### Does this PR introduce _any_ user-facing change?

No. The helper produces the same `SQLConf` the inline code did.

### How was this patch tested?

Behavior-preserving extraction, covered by existing SQL UDF resolution tests; CI on
this PR compiles and runs them.

### Was this patch authored or co-authored using generative AI tooling?

Generated-by: Claude Code (Anthropic Claude Opus)

Closes #56895 from daniel-lunin-db/SPARK-57756.

Authored-by: daniel-lunin-db <daniel.lunin@databricks.com>
Signed-off-by: Max Gekk <max.gekk@gmail.com>
(cherry picked from commit e40dc30)
Signed-off-by: Max Gekk <max.gekk@gmail.com>

Author: daniel-lunin-db

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala

@@ -327,6 +327,33 @@ object Analyzer {
       }
     }
   }
+
+  /**
+   * Builds the throwaway [[SQLConf]] used to resolve a SQL UDF body, seeded from the function's
+   * stored configs. Call sites share this seeding but differ in how they apply ANSI and session
+   * overrides, so both are parameters:
+   *  - `alwaysSetAnsiValue` calls [[trySetAnsiValue]] unconditionally (the SessionCatalog plan
+   *    builders); when false, ANSI is set only as part of the session overlay below (the
+   *    [[Analyzer#ResolveSQLFunctions]] / [[Analyzer#ResolveSQLTableFunctions]] rules).
+   *  - `applySessionOverrides` overlays the active session's retained resolution configs via
+   *    [[retainResolutionConfigsForAnalysis]], gated on
+   *    [[SQLConf.APPLY_SESSION_CONF_OVERRIDES_TO_FUNCTION_RESOLUTION]].
+   */
+  def buildSQLFunctionConf(
+      function: SQLFunction,
+      applySessionOverrides: Boolean,
+      alwaysSetAnsiValue: Boolean): SQLConf = {
+    val functionConf = new SQLConf()
+    function.getSQLConfigs.foreach { case (k, v) => functionConf.settings.put(k, v) }
+    if (alwaysSetAnsiValue) {
+      trySetAnsiValue(functionConf)
+    }
+    if (applySessionOverrides &&
+        conf.getConf(SQLConf.APPLY_SESSION_CONF_OVERRIDES_TO_FUNCTION_RESOLUTION)) {
+      retainResolutionConfigsForAnalysis(newConf = functionConf, existingConf = conf)
+    }
+    functionConf
+  }
 }
 
 /**
@@ -2700,11 +2727,10 @@ class Analyzer(
       val plan = v1SessionCatalog.makeSQLFunctionPlan(f.name, f.function, f.inputs)
       val resolved = SQLFunctionContext.withSQLFunction {
         // Resolve the SQL function plan using its context.
-        val newConf = new SQLConf()
-        f.function.getSQLConfigs.foreach { case (k, v) => newConf.settings.put(k, v) }
-        if (conf.getConf(SQLConf.APPLY_SESSION_CONF_OVERRIDES_TO_FUNCTION_RESOLUTION)) {
-          Analyzer.retainResolutionConfigsForAnalysis(newConf = newConf, existingConf = conf)
-        }
+        val newConf = Analyzer.buildSQLFunctionConf(
+          function = f.function,
+          applySessionOverrides = true,
+          alwaysSetAnsiValue = false)
         SQLConf.withExistingConf(newConf) {
           AnalysisContext.withAnalysisContext(f.function) {
             executeSameContext(plan)
@@ -3010,11 +3036,10 @@ class Analyzer(
       _.containsPattern(SQL_TABLE_FUNCTION)) {
       case SQLTableFunction(name, function, inputs, output) =>
         // Resolve the SQL table function plan using its function context.
-        val newConf = new SQLConf()
-        function.getSQLConfigs.foreach { case (k, v) => newConf.settings.put(k, v) }
-        if (conf.getConf(SQLConf.APPLY_SESSION_CONF_OVERRIDES_TO_FUNCTION_RESOLUTION)) {
-          Analyzer.retainResolutionConfigsForAnalysis(newConf = newConf, existingConf = conf)
-        }
+        val newConf = Analyzer.buildSQLFunctionConf(
+          function = function,
+          applySessionOverrides = true,
+          alwaysSetAnsiValue = false)
         val resolved = SQLConf.withExistingConf(newConf) {
           val plan = v1SessionCatalog.makeSQLTableFunctionPlan(name, function, inputs, output)
           SQLFunctionContext.withSQLFunction {

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/catalog/SessionCatalog.scala

@@ -1893,9 +1893,10 @@ class SessionCatalog(
     val funcName = function.name.funcName
 
     // Use captured SQL configs when parsing a SQL function.
-    val conf = new SQLConf()
-    function.getSQLConfigs.foreach { case (k, v) => conf.settings.put(k, v) }
-    Analyzer.trySetAnsiValue(conf)
+    val conf = Analyzer.buildSQLFunctionConf(
+      function = function,
+      applySessionOverrides = false,
+      alwaysSetAnsiValue = true)
     SQLConf.withExistingConf(conf) {
       val inputParam = function.inputParam
       val returnType = function.getScalarFuncReturnType