docs: dedupe precision definition and point to authoritative forbidden-patterns list

Author: sbryngelson

.claude/rules/fortran-conventions.md

@@ -53,10 +53,8 @@ Enforced by convention/code review (not automated):
 - Fortran-side runtime validation also exists in `m_checker*.fpp` files using `@:PROHIBIT`
 
 ## Precision Types
-- `wp` (working precision): used for computation. Double by default.
-- `stp` (storage precision): used for field data arrays and I/O. Double by default.
-- In single-precision mode (`--single`): both become single.
-- In mixed-precision mode (`--mixed`): wp=double, stp=half.
+`wp`/`stp` are defined in `CLAUDE.md` (`wp` = computation, `stp` = field-data storage + I/O). Detail:
+- Modes: default both double; `--single` → both single; `--mixed` → wp=double, stp=half.
 - MPI type matching: `mpi_p` must match `wp`, `mpi_io_p` must match `stp`.
 - Always use generic intrinsics: `sqrt` not `dsqrt`, `abs` not `dabs`.
 - Cast with `real(..., wp)` or `real(..., stp)`, never `dble(...)`.

CLAUDE.md

@@ -104,6 +104,7 @@ NEVER use double-precision intrinsics: `dsqrt`, `dexp`, `dlog`, `dble`, `dabs`,
 NEVER use `d` exponent literals (`1.0d0`). Use `1.0_wp` instead.
 NEVER use `stop` or `error stop`. Use `call s_mpi_abort()` or `@:PROHIBIT()`/`@:ASSERT()`.
 NEVER use `goto`, `COMMON` blocks, or global `save` variables.
+  (Headline subset; full lint-enforced list — incl. Python/shell rules — in `.claude/rules/fortran-conventions.md`.)
 
 Every `@:ALLOCATE(...)` MUST have a matching `@:DEALLOCATE(...)`.
 Every new parameter MUST be added in at least 2 places (3 if it has constraints):