feat(scala): parse indented while and end markers (semgrep#7700)

## What:
This PR makes it so we can parse `while` bodies that are indented, like:
```
while (true)
  val x = 2
  val y = 3
```

## Why:
SCALA 3

## How:
Just added the `opportunisticIndent`, similar to what we did for semgrep#7498 

## Test plan:
Included test, and parse rate `0.9806002679202549` ->
`0.9810696267622084`

PR checklist:

- [X] Purpose of the code is [evident to future
readers](https://semgrep.dev/docs/contributing/contributing-code/#explaining-code)
- [X] Tests included or PR comment includes a reproducible test plan
- [X] Documentation is up-to-date
- [ ] A changelog entry was [added to
changelog.d](https://semgrep.dev/docs/contributing/contributing-code/#adding-a-changelog-entry)
for any user-facing change
- [X] Change has no security implications (otherwise, ping security
team)

If you're unsure about any of this, please see:

- [Contribution
guidelines](https://semgrep.dev/docs/contributing/contributing-code)!
- [One of the more specific guides located
here](https://semgrep.dev/docs/contributing/contributing/)

languages/scala/ast/AST_scala.ml

@@ -356,6 +356,7 @@ and block_stat =
   | I of import
   | Ex of import
   | Ext of extension
+  | End of end_marker
   | E of expr
   (* just at the beginning of top_stat *)
   | Package of package
@@ -468,6 +469,11 @@ and definition_kind =
   (* class/traits/objects *)
   | Template of template_definition
 
+(*****************************************************************************)
+(* End Marker *)
+(*****************************************************************************)
+and end_marker = { end_tok : tok; end_kind : tok }
+
 (*****************************************************************************)
 (* Extensions *)
 (*****************************************************************************)

languages/scala/generic/scala_to_generic.ml

@@ -756,6 +756,9 @@ and v_block_stat x : G.item list =
   | E v1 ->
       let v1 = v_expr_for_stmt v1 in
       [ v1 ]
+  | End v1 ->
+      let v1 = v_end_marker v1 in
+      [ v1 ]
   | Ext v1 -> v_extension v1
   | Package v1 ->
       let ipak, ids = v_package v1 in
@@ -914,6 +917,9 @@ and v_given_definition { gsig; gkind } =
       G.OtherDef (todo_kind, v1 @ [ G.Anys v2 ]) );
   ]
 
+and v_end_marker { end_tok; end_kind } : G.stmt =
+  G.OtherStmt (OS_Todo, [ G.Tk end_tok; G.Tk end_kind ]) |> G.s
+
 and v_extension { ext_tok = _; ext_tparams; ext_using; ext_param; ext_methods }
     : G.stmt list =
   let tparams =

languages/scala/recursive_descent/Parser_scala_recursive_descent.ml

@@ -2488,6 +2488,7 @@ and parseWhile in_ : stmt =
         accept (Kdo ab) in_;
         fb (Tok.unsafe_fake_tok "") e
   in
+  opportunisticIndent in_;
   let body = expr in_ in
   (* ast: makeWhile(cond, body) *)
   While (ii, cond, body)
@@ -3651,6 +3652,27 @@ let localDef implicitMod in_ : definition =
          defs
          (* AST: if RBRACE | CASE defs :+ literalUnit *))
 
+(* ------------------------------------------------------------------------- *)
+(* End Marker *)
+(* ------------------------------------------------------------------------- *)
+
+(** {{{
+ *  EndMarker         ::=  ‘end’ EndMarkerTag    -- when followed by EOL
+ *  EndMarkerTag      ::=  id | ‘if’ | ‘while’ | ‘for’ | ‘match’ | ‘try’
+                        |  ‘new’ | ‘this’ | ‘given’ | ‘extension’ | ‘val’
+ *  }}}
+*)
+
+let endMarker in_ : end_marker =
+  let end_tok = TH.info_of_tok in_.token in
+  accept (ID_LOWER ("end", ab)) in_;
+  let end_kind = TH.info_of_tok in_.token in
+  (* We could dispatch on the various cases, but every single case ends up
+     with some kind of token. So let's just say that it's a single token.
+  *)
+  skipToken in_;
+  { end_tok; end_kind }
+
 (* ------------------------------------------------------------------------- *)
 (* Extension *)
 (* ------------------------------------------------------------------------- *)
@@ -3777,6 +3799,7 @@ let statSeq ?(errorMsg = "illegal start of definition") ?(rev = false) stat in_
   *                 | Annotations [implicit] [lazy] Def
   *                 | Annotations LocalModifiers TmplDef
   *                 | Expr1
+  *                 | EndMarker
   *                 | Extension
   *                 |
   *  }}}
@@ -3795,6 +3818,9 @@ let blockStatSeqInner in_ : top_stat option =
     if not (TH.isCaseDefEnd in_.token) then acceptStatSepOpt in_
   in
   match in_.token with
+  | ID_LOWER ("end", _) ->
+      let x = endMarker in_ in
+      Some (End x)
   | ID_LOWER ("extension", _) ->
       let x = extension in_ in
       Some (Ext x)
@@ -3915,6 +3941,7 @@ let indentedExprOrBlockStatSeqUntil ?(until = None) in_ =
  *                     | Annotations Modifiers Dcl
  *                     | Expr1
  *                     | super ArgumentExprs {ArgumentExprs}
+ *                     | EndMarker
  *                     | Extension
  *                     |
  *                     | Annotations Modifiers EnumCase
@@ -3929,6 +3956,9 @@ let templateStat in_ : template_stat option =
   | Kexport _ ->
       let x = exportClause in_ in
       Some (Ex x)
+  | ID_LOWER ("end", _) ->
+      let x = endMarker in_ in
+      Some (End x)
   | ID_LOWER ("extension", _) ->
       let x = extension in_ in
       Some (Ext x)
@@ -3953,6 +3983,7 @@ let templateStats in_ : template_stat list = statSeq templateStat in_
  *            | package object ObjectDef
  *            | Import
  *            | Export
+ *            | EndMarker
  *            | Extension
  *            |
  *  }}}
@@ -3969,6 +4000,9 @@ let topStat in_ : top_stat option =
   | Kexport _ ->
       let x = exportClause in_ in
       Some (Ex x)
+  | ID_LOWER ("end", _) ->
+      let x = endMarker in_ in
+      Some (End x)
   | ID_LOWER ("extension", _) ->
       let x = extension in_ in
       Some (Ext x)
@@ -4011,7 +4045,12 @@ let templateStatSeq ~isPre in_ : self_type option * block =
     if
       TH.isExprIntro in_.token
       && (not (is_modifier in_))
-      && not (in_.token =~= ID_LOWER ("extension", ab))
+      (* We add this here, because there are some "soft" modifiers that can start
+         a templateIntro, such as "extension", "end", etc.
+         If we did not do this, template stats that start with those would always
+         enter this expression case, even though they denote something else.
+      *)
+      && not (TH.isTemplateIntro in_.token)
     then (
       let first = expr ~location:InTemplate in_ in
       match in_.token with

languages/scala/recursive_descent/Token_helpers_scala.ml

@@ -394,6 +394,7 @@ let isTemplateIntro = function
   | Kenum _
   | Ktrait _
   | ID_LOWER ("given", _)
+  | ID_LOWER ("end", _)
   | ID_LOWER ("extension", _) ->
       true
   (*TODO | Kcaseobject | | Kcaseclass *)

tests/parsing/scala/end_markers.scala

@@ -0,0 +1,10 @@
+
+object A:
+  extension (a: File)
+    inline def x: String = "hi" 
+  end extension
+
+  def foo(x: T) = x
+  end def
+
+end foo

tests/parsing/scala/indented_while.scala

@@ -0,0 +1,4 @@
+
+while (true)
+  val x = 2
+  val y = 3