expose max features for normal solver, and improve doc

mllib/src/main/scala/org/apache/spark/ml/optim/NormalEquationSolver.scala

@@ -157,7 +157,7 @@ private[ml] class QuasiNewtonSolver(
  * Exception thrown when solving a linear system Ax = b for which the matrix A is non-invertible
  * (singular).
  */
-private[ml] class SingularMatrixException(message: String, cause: Throwable)
+private[spark] class SingularMatrixException(message: String, cause: Throwable)
   extends IllegalArgumentException(message, cause) {
 
   def this(message: String) = this(message, null)

mllib/src/main/scala/org/apache/spark/ml/regression/LinearRegression.scala

@@ -160,14 +160,13 @@ class LinearRegression @Since("1.3.0") (@Since("1.3.0") override val uid: String
   /**
    * Set the solver algorithm used for optimization.
    * In case of linear regression, this can be "l-bfgs", "normal" and "auto".
-   * "l-bfgs" denotes Limited-memory BFGS which is a limited-memory quasi-Newton
-   * optimization method. "normal" denotes using Normal Equation as an analytical
-   * solution to the linear regression problem.
-   * The default value is "auto" which means that the solver algorithm is
-   * selected automatically.
-   *
-   * The Normal Equation solver is limited to a few thousand features; when needed,
-   * [[LinearRegression]] will automatically fall back to iterative optimization methods.
+   *  - "l-bfgs" denotes Limited-memory BFGS which is a limited-memory quasi-Newton
+   *    optimization method.
+   *  - "normal" denotes using Normal Equation as an analytical solution to the linear regression
+   *    problem.  This solver is limited to [[LinearRegression.MAX_FEATURES_FOR_NORMAL_SOLVER]].
+   *  - "auto" (default) means that the solver algorithm is selected automatically.
+   *    The Normal Equations solver will be used when possible, but this will automatically fall
+   *    back to iterative optimization methods when needed.
    *
    * @group setParam
    */
@@ -407,6 +406,13 @@ object LinearRegression extends DefaultParamsReadable[LinearRegression] {
 
   @Since("1.6.0")
   override def load(path: String): LinearRegression = super.load(path)
+
+  /**
+   * When using [[LinearRegression.solver]] == "normal", the solver must limit the number of
+   * features to at most this number.
+   */
+  @Since("2.1.0")
+  val MAX_FEATURES_FOR_NORMAL_SOLVER: Int = WeightedLeastSquares.MAX_NUM_FEATURES
 }
 
 /**