Transition all relevant testcases to it. In the process, port
docstrings/comments from test.py files to expected_concrete_syntax.lkt
(now test.lkt) sources and fix stylechecks issues there.
(cherry picked from commit e1c4fe5804)
This makes the convention consistent with the DSL, and avoids
workarounds for conflicts with Lkt keywords: in Libadalang, the Null
token can stay Null, instead of null_tok (no API breakage needed).
This test fails sometimes because it makes no sense at all. There is no
information about rewriting context being equivalent when closing and
reopening a new one (see doc for 'langkit.rewriting.start_rewriting').
Before this commit, node typed structure fields were stored as is in
Java structure wrapping classes. Entities were 'Entity' and bare node
were 'PointerWrapper'.
Now, those fields are stored as '<my_lang>Node' in their Java wrapping
classes.
That'll allow us to go from (thin_symbol, sym_table) to symbol, and vice
versa, where with the old representation there was no other way to go
from symbol to thin than redoing a full-lookup.
That'll help us transition to thin-symbol storage everywhere.
Before this commit, in the 'AnalysisUnit.equals' method, the other value
was compared with itself, resulting in the method returning always true.
Now 'other' is compared to 'this'.
