Remove support for Transactional Memory Extension (TME)

The Transactional Memory Extension (TME) was introduced as part of
Armv9-A but has not been adopted by the ecosystem. This mirrors what
Arm has observed with similar extensions in other architectures.

Support for TME has now been officially withdrawn, as noted here:

```
 FEAT_TME is withdrawn from all future versions of Arm®
 Architecture Reference Manual for A-profile architecture.
```

referenced in Known Issue D24093, documented here:
  https://developer.arm.com/documentation/102105/lb-05/

Author: jthackray

main/acle.md

@@ -467,6 +467,7 @@ Armv8.4-A [[ARMARMv84]](#ARMARMv84). Support is added for the Dot Product intrin
 * Added feature test macro for FEAT_CSSC.
 * Added support for modal 8-bit floating point matrix multiply-accumulate widening intrinsics.
 * Added support for 16-bit floating point matrix multiply-accumulate widening intrinsics.
+* Removed all references to Transactional Memory Extension (TME)
 
 ### References
 
@@ -1769,13 +1770,6 @@ the Armv8.1-A [[ARMARMv81]](#ARMARMv81) architecture are supported on this targe
 Note: It is strongly recommended that standardized C11/C++11 atomics are used to
 implement atomic operations in user code.
 
-### Transactional Memory Extension
-
-`__ARM_FEATURE_TME` is defined to `1` if the Transactional Memory
-Extension instructions are supported in hardware and intrinsics defined
-in [Transactional Memory Extension (TME)
-intrinsics](#transactional-memory-extension-tme-intrinsics) are available.
-
 ### Armv8.7-A Load/Store 64 Byte extension
 
 `__ARM_FEATURE_LS64` is defined to 1 if the Armv8.7-A `LD64B`,
@@ -14486,40 +14480,6 @@ The MVE load and store instructions provide for alignment assertions, which may
 speed up access to aligned data (and will fault access to unaligned data).  The
 MVE intrinsics do not directly provide a means for asserting alignment.
 
-# Transactional Memory Extension (TME) intrinsics
-
-## Introduction
-
-This section describes the intrinsics for the instructions of the
-Transactional Memory Extension (TME). TME adds support for transactional
-execution where transactions are started and
-committed by a set of new instructions. The TME instructions are present
-in the AArch64 execution state only.
-
-TME is designed to improve performance in cases where larger system scaling
-requires atomic and isolated access to data structures whose composition is
-dynamic in nature and therefore not readily amenable to fine-grained locking
-or lock-free approaches.
-
-TME transactions are *isolated*. This means that transactional stores are
-hidden from other observers, and transactional loads cannot see stores from
-other observers until the transaction commits. Also, if the transaction fails
-then stores to memory and writes to registers by the transaction are discarded
-and the processor returns to the state it had when the transaction started.
-
-TME transactions are *best-effort*. This means that the architecture does not
-guarantee success for any transaction. The architecture requires that all
-transactions specify a failure handler allowing the software to fallback to a
-non-transactional alternative to provide guarantees of forward progress.
-
-TME defines *flattened nesting* of transactions, where nested transactions are
-subsumed by the outer transaction. This means that the effects of a nested
-transaction do not become visible to other observers until the outer
-transaction commits. When a nested transaction fails it causes the
-outer transaction, and all nested transactions within, to fail.
-
-The TME intrinsics are available when `__ARM_FEATURE_TME` is defined.
-
 ## Failure definitions
 
 Transactions can fail due to various causes. The following macros