fix semantic highlighting on names that are wrapped in parentheses across multiple lines

[DetachHead]

fixes #353

[github-actions]

Diff from mypy_primer, showing the effect of this PR on open source code:

sympy (https://github.com/sympy/sympy)
-     Type of "n" is "Unknown | ((n: int = 15, subs: dict[Basic, Basic | float] | None = None, maxn: int = 100, chop: bool = False, strict: bool = False, quad: str | None = None, verbose: bool = False) -> Expr)" (reportUnknownMemberType)
+     Type of "n" is "Unknown | ((n: int = 15, subs: dict[Basic, Basic | float] | None = None, maxn: int = 100, chop: bool = False, strict: bool = False, quad: str | None = None, verbose: bool = False) -> Expr) | Any" (reportUnknownMemberType)
-   /tmp/mypy_primer/projects/sympy/sympy/solvers/tests/test_solveset.py:506:20 - warning: Argument type is unknown
-     Argument corresponds to parameter "args" in function "__new__" (reportUnknownArgumentType)
+   /tmp/mypy_primer/projects/sympy/sympy/solvers/tests/test_solveset.py:706:26 - warning: Argument type is Any
+     Argument corresponds to parameter "f" in function "solveset_real" (reportAny)
-     Type of "expr1" is "Equality | Unknown | Relational | Unequality" (reportUnknownVariableType)
+     Type of "expr1" is "Equality | Unknown" (reportUnknownVariableType)
-     Type of "expr2" is "Equality | Unknown | Relational | Unequality" (reportUnknownVariableType)
+     Type of "expr2" is "Equality | Unknown" (reportUnknownVariableType)
-     Type of "eqns2" is "list[Equality | Unknown | Relational | Unequality | Any]" (reportUnknownVariableType)
+     Type of "eqns2" is "list[Equality | Unknown | Any]" (reportUnknownVariableType)
-     Type of "number" is "Unknown | One | NegativeOne | Zero | Integer | Rational | NaN | ComplexInfinity | Half | Expr" (reportUnknownVariableType)
+     Type of "number" is "Unknown | One | NegativeOne | Zero | Integer | Rational | NaN | ComplexInfinity | Half | Infinity | NegativeInfinity | Float | _NotImplementedType | Expr" (reportUnknownVariableType)
+   /tmp/mypy_primer/projects/sympy/sympy/solvers/tests/test_solveset.py:2793:21 - warning: Argument type is Any
+     Argument corresponds to parameter "f" in function "solveset" (reportAny)
+   /tmp/mypy_primer/projects/sympy/sympy/solvers/tests/test_solveset.py:3411:13 - warning: Argument type is Any
+     Argument corresponds to parameter "lhs" in function "__new__" (reportAny)
-     Type of "eq1" is "list[Equality | Unknown | Relational | Unequality]" (reportUnknownVariableType)
+     Type of "eq1" is "list[Equality | Unknown]" (reportUnknownVariableType)
-     Type of "eq2" is "list[Equality | Unknown | Relational | Unequality]" (reportUnknownVariableType)
+     Type of "eq2" is "list[Equality | Unknown]" (reportUnknownVariableType)
-     Type of "eqs_eq" is "list[Equality | Unknown | Relational | Unequality]" (reportUnknownVariableType)
+     Type of "eqs_eq" is "list[Equality | Unknown]" (reportUnknownVariableType)
-     Type of "eq" is "Equality | Unknown | Relational | Unequality" (reportUnknownVariableType)
+     Type of "eq" is "Equality | Unknown" (reportUnknownVariableType)
-   /tmp/mypy_primer/projects/sympy/sympy/solvers/tests/test_solveset.py:3513:5 - warning: Type of "e" is partially unknown
-     Type of "e" is "ComplexInfinity | Rational | NaN | One | NegativeOne | Zero | Integer | Half | Unknown" (reportUnknownVariableType)
-     Type of "ans" is "set[tuple[Any | Unknown, Any] | tuple[Unknown, Any]]" (reportUnknownVariableType)
+     Type of "ans" is "set[tuple[Any, Any] | tuple[Unknown, Any]]" (reportUnknownVariableType)
-     Type of "P" is "(condition: Unknown, given_condition: Unknown | None = None, numsamples: Unknown | None = None, evaluate: bool = True, **kwargs: Unknown) -> (Unknown | Any | BernoulliDistribution | Probability | Zero | One | Self@Expr | tuple[Unknown, ...] | Sum | Expr | NaN | Piecewise | Basic | Equality | Relational | Unequality | ComplexInfinity | Float | Half | Infinity | Integer | Lambda | NegativeInfinity | NegativeOne | Number | Rational | int | None)" (reportUnknownVariableType)
+     Type of "P" is "(condition: Unknown, given_condition: Unknown | None = None, numsamples: Unknown | None = None, evaluate: bool = True, **kwargs: Unknown) -> (Unknown | Any | BernoulliDistribution | Probability | Zero | One | Expr | Float | ComplexInfinity | Rational | NaN | NegativeOne | Integer | Half | Infinity | NegativeInfinity | _NotImplementedType | Self@Expr | tuple[Unknown, ...] | Sum | Piecewise | Basic | Equality | Lambda | Number | int | None)" (reportUnknownVariableType)
-     Type of "E" is "(expr: Unknown, condition: Unknown | None = None, numsamples: Unknown | None = None, evaluate: bool = True, **kwargs: Unknown) -> (Unknown | Basic | Expectation | Expr | tuple[Unknown, ...] | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Relational | Unequality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
+     Type of "E" is "(expr: Unknown, condition: Unknown | None = None, numsamples: Unknown | None = None, evaluate: bool = True, **kwargs: Unknown) -> (Unknown | Basic | Expectation | Expr | tuple[Unknown, ...] | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
-     Type of "H" is "(expr: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (int | Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Relational | Unequality | Integral | ExpectationMatrix | None)" (reportUnknownVariableType)
+     Type of "H" is "(expr: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (int | Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Integral | ExpectationMatrix | None)" (reportUnknownVariableType)
-     Type of "covariance" is "(X: Unknown, Y: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (CrossCovarianceMatrix | Covariance | Unknown | Basic | Expectation | Expr | tuple[Unknown, ...] | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Relational | Unequality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
+     Type of "covariance" is "(X: Unknown, Y: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (CrossCovarianceMatrix | Covariance | Unknown | Basic | Expectation | Expr | tuple[Unknown, ...] | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
-     Type of "entropy" is "(expr: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (int | Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Relational | Unequality | Integral | ExpectationMatrix | None)" (reportUnknownVariableType)
+     Type of "entropy" is "(expr: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (int | Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Integral | ExpectationMatrix | None)" (reportUnknownVariableType)
-     Type of "correlation" is "(X: Unknown, Y: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Unknown | Expr | Any | ComplexInfinity | Rational | NaN | One | NegativeOne | Zero | Integer | Half)" (reportUnknownVariableType)
+     Type of "correlation" is "(X: Unknown, Y: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Unknown | Expr | Any | ComplexInfinity | Rational | NaN | One | NegativeOne | Zero | Integer | Half | Infinity | NegativeInfinity | Float | _NotImplementedType)" (reportUnknownVariableType)
-     Type of "factorial_moment" is "(X: Unknown, n: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Relational | Unequality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
+     Type of "factorial_moment" is "(X: Unknown, n: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Expr | Basic | Expectation | tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Any | Zero | NaN | Piecewise | Equality | Integral | ExpectationMatrix | Literal[0] | None)" (reportUnknownVariableType)
-     Type of "coskewness" is "(X: Unknown, Y: Unknown, Z: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Unknown | Expr | Any | ComplexInfinity | Rational | NaN | One | NegativeOne | Zero | Integer | Half)" (reportUnknownVariableType)
+     Type of "coskewness" is "(X: Unknown, Y: Unknown, Z: Unknown, condition: Unknown | None = None, **kwargs: Unknown) -> (Unknown | Expr | Any | ComplexInfinity | Rational | NaN | One | NegativeOne | Zero | Integer | Half | Infinity | NegativeInfinity | Float | _NotImplementedType)" (reportUnknownVariableType)
-   /tmp/mypy_primer/projects/sympy/sympy/stats/compound_rv.py:107:9 - warning: Return type, "tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Expr | Any | Zero | NaN | Piecewise | Basic | Equality | Relational | Unequality | Integral | Literal[0] | None", is partially unknown (reportUnknownParameterType)
+   /tmp/mypy_primer/projects/sympy/sympy/stats/compound_rv.py:107:9 - warning: Return type, "tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Expr | Any | Zero | NaN | Piecewise | Basic | Equality | Integral | Literal[0] | None", is partially unknown (reportUnknownParameterType)
-     Type of "compute_expectation" is "(expr: Unknown, rvs: Unknown | None = None, **kwargs: Unknown) -> (tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Expr | Any | Zero | NaN | Piecewise | Basic | Equality | Relational | Unequality | Literal[0] | None)" (reportUnknownMemberType)
+     Type of "compute_expectation" is "(expr: Unknown, rvs: Unknown | None = None, **kwargs: Unknown) -> (tuple[Unknown, ...] | Unknown | Sum | Self@Expr | Expr | Any | Zero | NaN | Piecewise | Basic | Equality | Literal[0] | None)" (reportUnknownMemberType)

... (truncated 4073 lines) ...

arviz (https://github.com/arviz-devs/arviz): 3.83x slower (8.2s -> 31.6s in a single noisy sample)