chore: repin Cargo.Bazel.json.lock

The cherry-pick of #11090 ("fix: patch pcre2-sys for reproducibility")
carried over master's `Cargo.Bazel.json.lock`, whose `checksum` is a
digest over the crate_universe config and splicing manifest — i.e. over
`bazel/rust.MODULE.bazel` plus every workspace `Cargo.toml`. Those differ
between master and this RC branch (master has since bumped `comparable`
to the dfinity fork, added `axum-otel-metrics`, `duration-string`,
`exitcode`, bumped `canbench-rs` and `ciborium`, …), so the digest is
stale here and every Bazel invocation fails at analysis time:

    Error: Digests do not match:
      Current  Digest("987c09bb853448361351d6b167b9e528c23cea8fdaa95ce28b65cc3a9114fbce")
      != Expected Digest("4991a9aa71efd82294bf1844c7f4e677fc0fdad5c5d280d67766a7edd824e45b")
    The current `lockfile` is out of date for 'crate_index'.

Regenerated with `./bin/bazel-pin.sh`. The resolved crate set was already
correct for this branch, so the only change is the recorded checksum,
which now matches the digest CI computes. `bazel query @crate_index//:all`
passes again.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

chore: repin Cargo.Bazel.json.lock

The cherry-pick of #11090 ("fix: patch pcre2-sys for reproducibility")
carried over master's `Cargo.Bazel.json.lock`, whose `checksum` is a
digest over the crate_universe config and splicing manifest — i.e. over
`bazel/rust.MODULE.bazel` plus every workspace `Cargo.toml`. Those differ
between master and this RC branch, which is 28 commits behind (master has
since moved `comparable` to the dfinity fork, added `axum-otel-metrics`,
`duration-string` and `exitcode`, bumped `canbench-rs` and `ciborium`, …),
so the digest is stale here and every Bazel invocation fails at analysis
time:

    Error: Digests do not match:
      Current  Digest("987c09bb853448361351d6b167b9e528c23cea8fdaa95ce28b65cc3a9114fbce")
      != Expected Digest("4991a9aa71efd82294bf1844c7f4e677fc0fdad5c5d280d67766a7edd824e45b")
    The current `lockfile` is out of date for 'crate_index'.

Regenerated with `./bin/bazel-pin.sh`. The resolved crate set was already
correct for this branch, so the only change is the recorded checksum,
which now matches the digest CI computes. `bazel query @crate_index//:all`
and `bazel build --nobuild //publish/canisters:bundle` pass again.

Same fix as the one applied to rc--2026-08-06_04-21-revert_dmt-fix-repro.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

reverts d21b8e8

feat(registry): Enable blank replica_version_id for engines. (#10888)

This is "just" a flag flip, but it suddenly makes a whole bunch of code
active.

# Future Work

Existing engines cannot yet transition to blank replica_version_id. That
will be fixed in an upcoming PR.

# References

[👈 Previous PR][prev]

[prev]: #10885

---------

Co-authored-by: Daniel Wong <daniel.wong@dfinity.org>

reverts d21b8e8

feat(Core): Switch to deterministic memory tracker (#10789)

This PR unifies page instruction charging. The deterministic memory
tracker (DMT) now charges for all combinations of {heap, stable} x
{read, write} page access types. The cost is uniform and increased to
5000 instructions per OS page (4KiB).

Unchanged is the additional cost of copying dirty pages after the
message execution (3000 instructions per page).

So overall, the costs change like this:
```
               previously     dmt
heap read:              0    5000
heap write:         *1000   *5000
stable read:            0    5000
stable write:       *1000   *5000
```

* additionally charged for copy overhead after the message (+3000)

The dirty stable pages tracking via the injected system API replacements
is changed: The instruction counter is no longer decreased, because the
DMT does it automatically. But the bitmap tracking of stable pages is
kept, because it's still needed to limit overall stable memory access.

As a consequence of this change, heap accesses are charged
deterministically and immediately (during the message/slice, rather than
afterwards), which means that messages with lots of heap accesses are
DTS'd properly. While this incurs additional cycle costs, it will allow
us to increase the Wasm heap size to much more than the current 6GiB
limit.

Squashed from #10536

---------

Co-authored-by: Michael Weigelt <michael.weigelt@dfinity.org>
Co-authored-by: michael-weigelt <122277901+michael-weigelt@users.noreply.github.com>

bump libflate

(cherry picked from commit 5545c50)

bump libflate

(cherry picked from commit 5545c50)

This PR unifies page instruction charging. The deterministic memory t…

…racker (DMT) now charges for all combinations of {heap, stable} x {read, write} page access types. The cost is uniform and increased to 5000 instructions per OS page (4KiB).

Unchanged is the additional cost of copying dirty pages after the message execution (3000 instructions per page).

So overall, the costs change like this:

```
               previously     dmt
heap read:              0    5000
heap write:         *1000   *5000
stable read:            0    5000
stable write:       *1000   *5000

    *additionally charged for copy overhead after the message (+3000)
```

The dirty stable pages tracking via the injected system API replacements is changed: The instruction counter is no longer decreased, because the DMT does it automatically. But the bitmap tracking of stable pages is kept, because it's still needed to limit overall stable memory access.

As a consequence of this change, heap accesses are charged deterministically and immediately (during the message/slice, rather than afterwards), which means that messages with lots of heap accesses are DTS'd properly. While this incurs additional cycle costs, it will allow us to increase the Wasm heap size to much more than the current 6GiB limit.

chore: Add `cost_http_request_v2` parameter sanity check (#10866)

We allow an increase of up to 2x in size, in order to enable future
parameter extensions without breaking backward compatibility.