devel/llvm{19,20,21}: fix build on riscv64

=======================<phase: check-sanity   >============================
====> You must select one and only one option from the BACKENDS single
=====> No option was selected (and one must be)

Approved by:	portmgr (build fix blanket)
MFH:		2025Q3

python.mk: rename PYTHON_EXT_SUFFIX to PYTHON_TAG, document, etc

EXT_SUFFIX, according to PEP 3149, refers to the full tag and
extension for compiled extension module objects, eg .cpython-311.so,
.cpython-313t.so, etc. We do not use the correct semantic meaning,
and this usage becomes ambiguous and confusing when tags do not
match between bytecode and compiled extension module objects.

Rename our PYTHON_EXT_SUFFIX to PYTHON_TAG to align with PEP 3147's
specification of a magic tag, which consists of implementation name
and shorthand version only. This is meant for bytecode and other
files containing this tag in their filename that do not depend on
a specific Python ABI for the same version. Chase all existing
consumers.

devel/llvm19: reenable AMDGPU and WASM on i386 by default

This is needed (among other reasons) to support wine because wine is
mostly 32-bit.

Requested by:	manu
Sponsored by:	DARPA, AFRL

devel/llvm19: 19.1.7 release

See the release announcement for changes:
    https://discourse.llvm.org/t/llvm-19-1-7-released/84062

Sponsored by:	DARPA, AFRL

devel/llvm19: Fix !CLANG plist and OPENMP

llvm-debuginfo-analyzer isn't part of Clang, but was matched by the
overly broad pattern `-analyzer`. [0]

Fix USE_LDCONFIG for OPENMP. [1]

Reported by:	jrtc27 [0], tijl [1]
Sponsored by:	DARPA, AFRL

devel/llvm*: flip LICENSE_FILE_BSD3CLAUSE setting [NFC]

The location of this license file moved in LLVM 16.  Since I'm doing
cleanup in this area and the split of ports is about 50/50, switch to
setting the old location in the older ports so the newer ports set fewer
variables.

Sponsored by:	DARPA, AFRL

devel/llvm*: LICENSE_DIR simplification

WRKSRC moved to the top of the source tree with llvm12.  As a result,
LICENSE_DIR became ${WRKSRC}/llvm.  Given that llvm11 the only port that
wasn't setting LICENSE_DIR, flip the script and default to
${WRKSRC}/llvm with llvm11 overriding.

Sponsored by:	DARPA, AFRL

devel/llvm*: don't define the ARM license [NFC]

All LLVM ports excluded it from the list as it was removed many years
ago.

Sponsored by:	DARPA, AFRL

devel/llvm*: Always define LLVM2 license [NFC]

All LLVM ports in the tree use the Apache License 2.0 with LLVM
Exceptions license so cerntralize the definition.

Sponsored by:	DARPA, AFRL

devel/llvm19: 19.1.6 release

See the release annoucement for details:
    https://discourse.llvm.org/t/llvm-19-1-6-released/83723

Backport loongarch64 triple support from upstream LLVM.

devel/llvm19: 19.1.5

See release annoucement for changes:
    https://discourse.llvm.org/t/llvm-19-1-5-released/83455

Sponsored by:	DARPA, AFRL

devel/llvm19: 19.1.4 release

See the release notes for changes:
    https://discourse.llvm.org/t/llvm-19-1-4-released/83230

Sponsored by:	DARPA, AFRL

devel/llvm19: 19.1.3 release

See the release annoucement for changes:
	https://discourse.llvm.org/t/llvm-19-1-3-relased/82829

Sponsored by:	DARPA, AFRL

devel/llvm19: 19.1.2 release

See the release annoucement for further details:

https://discourse.llvm.org/t/llvm-19-1-2-released/82550

devel/llvm19: 19.1.1 release

Remove obsolete build conflict with cmark.

PR:		239636, 274848

devel/llvm19: 19.1.0 release

Sponsored by: 	DARPA, AFRL

devel/llvm[789]: restore arm_bf16.h to ARM backend

arm_bf16.h is generated for both armv7 and aarch64 so move it back to
ARM.  We don't support AArch64 without ARM so we can just move it and
not need to worry about duplicate supression (I'd previously fixed the
issue in llvm19 in a more complicated manner so this change removes the
duplicate supression and documents why it's not needed.)

PR:		281179
Reported by:	Mark Millard

devel/llvm19: update to 19.1.0-rc4

arm_bf16.h is installed by the ARM backend as well as AArch64 so handle
that.

Patch out upstream 6d7e428df611 which traded a false negative for a
false positive when compiling compiler-rt for i386.

Fix default options for 32-bit powerpc arches.

PR:		281179

devel/llvm19: 19.1.0-rc3

Sponsored by:	DARPA, AFRL

devel/llvm19: prune build on 32-bit archs

Given trends in compiler resource use, it is incresingly implausible to
consider cross-development from a 32-bit system.  As such, stop spending
time building LLVM bits are are vinishingly unlikely to be used on
32-bit systems.

Default to BE_NATIVE (no cross build support) on 32-bit plaforms.
Disable BE_AMDGPU and BE_WASM as well.  BE_AMDGPU consumes enormous
amounts of memory.  BE_WASM is less problematic, but no one has stood up
to say they need it so turn it off for now.

Completely disable MLIR and POLLY on 32-bit.  Just building MLIR fails
routinly on armv7 and there aren't a lot of direct users (it's used by
FLANG, but FLANG is 64-bit only).  Polly is pretty niche and adds quite
a bit of build time.

Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D46239

devel/llvm19: 19.1.0-rc2

See https://discourse.llvm.org/t/llvm-19-1-0-rc2-released/80549 for a
change log.

Sponsored by:	 DARPA, AFRL

devel/llvm19: Update from llvm18 and connect

devel/llvm19: copy from devel/llvm18

This is a complete unmodifed copy to simplify history tracking.