…ributes (envoyproxy#1372)

Automatic merge from submit-queue.

Support Mixer HTTP filter to report sent.bytes and received.bytes attributes

**What this PR does / why we need it**:Support Mixer HTTP filter to send attributes "sent.bytes" and "received.bytes" in Report() calls. "sent.bytes" records total response sent in bytes, including response headers, body, and trailers. "received.bytes" records total request received in bytes, including request headers, body, and trailers.

**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes envoyproxy#1370 

**Special notes for your reviewer**:

**Release note**:
```release-note
NONE
```

…1878)

**Description**

This commit adds a translator that will convert a request sent to
`/anthropic/v1/messages` and `/v1/messages` endpoints for OpenAI schema
backends. It does not matter whether the OpenAI schema backend natively
supports the endpoint (e.g. vLLM) as translating should be a light/fast
enough process. This approach is also more versatile and future-proof
than just passing through the Anthropic Message Request to a backend
that natively supports it. It also follows the already-existing
structure for adding translators, path processor factories, and schema
translation.

A major example use case would be using AI Gateway to route requests
from Claude Code to several AI backends like locally hosted vLLM models
with LoRA adapters.

NOTE: vLLM is only used for local testing as I do not have access to
compute. The intended goal for this PR is to support any OpenAI
compatible backend/services using an Anthropic interface.

**Related Issues/PRs (if applicable)**

Fixes #1372 
Fixes #1867 

**Special notes for reviewers (if applicable)**
Claude Code was used to write most of the tests but were verified. It
would also be nice if the maintainers could review the other PR #1843 as
some of the Anthropic apischema here can be updated once #1843 is
merged.

<details>

<summary>Functional Test Results</summary>

Test for anthropic endpoints for OpenAI schema backends that natively
support it:
```
$ curl -v http://localhost:8080/v1/messages   -H "Content-Type: application/json"   -d '{
    "model": "Qwen/Qwen2.5-0.5B-Instruct",
    "messages": [
      {"role": "user", "content": "Say hello!"}
    ],
    "max_tokens": 100
  }'
* Host localhost:8080 was resolved.
* IPv6: ::1
* IPv4: 127.0.0.1
*   Trying [::1]:8080...
* Connected to localhost (::1) port 8080
> POST /v1/messages HTTP/1.1
> Host: localhost:8080
> User-Agent: curl/8.5.0
> Accept: */*
> Content-Type: application/json
> Content-Length: 143
> 
< HTTP/1.1 200 OK
< date: Fri, 20 Feb 2026 18:46:04 GMT
< server: uvicorn
< content-type: application/json
< content-length: 331
< 
* Connection #0 to host localhost left intact
{"id":"chatcmpl-36ec5b3d-4273-41e9-966b-ed742f7a93d1","type":"message","role":"assistant","content":[{"type":"text","text":"Hello! How can I assist you today?"}],"model":"Qwen/Qwen2.5-0.5B-Instruct","stop_reason":"end_turn","usage":{"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"input_tokens":32,"output_tokens":10}}
```
```
$ curl -v http://localhost:8080/anthropic/v1/messages   -H "Content-Type: application/json"   -d '{
    "model": "Qwen/Qwen2.5-0.5B-Instruct",
    "messages": [
      {"role": "user", "content": "Say hello!"}
    ],
    "max_tokens": 100
  }'
* Host localhost:8080 was resolved.
* IPv6: ::1
* IPv4: 127.0.0.1
*   Trying [::1]:8080...
* Connected to localhost (::1) port 8080
> POST /anthropic/v1/messages HTTP/1.1
> Host: localhost:8080
> User-Agent: curl/8.5.0
> Accept: */*
> Content-Type: application/json
> Content-Length: 143
> 
< HTTP/1.1 200 OK
< date: Fri, 20 Feb 2026 18:46:44 GMT
< server: uvicorn
< content-type: application/json
< content-length: 331
< 
* Connection #0 to host localhost left intact
{"id":"chatcmpl-f639ff32-4f89-48c5-b5b1-56878e641da6","type":"message","role":"assistant","content":[{"type":"text","text":"Hello! How can I assist you today?"}],"model":"Qwen/Qwen2.5-0.5B-Instruct","stop_reason":"end_turn","usage":{"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"input_tokens":32,"output_tokens":10}}
```

Port Forward logs
```
$ kubectl port-forward -n envoy-gateway-system svc/$ENVOY_SERVICE 8080:80
Forwarding from 127.0.0.1:8080 -> 10080
Forwarding from [::1]:8080 -> 10080
Handling connection for 8080
Handling connection for 8080
Handling connection for 8080
```

vLLM Logs (for both requests)
```
(APIServer pid=141923) INFO:     Started server process [141923]
(APIServer pid=141923) INFO:     Waiting for application startup.
(APIServer pid=141923) INFO:     Application startup complete.
(APIServer pid=141923) INFO:     172.18.0.2:46854 - "POST /v1/chat/completions HTTP/1.1" 200 OK
(APIServer pid=141923) INFO 02-20 13:46:05 [loggers.py:257] Engine 000: Avg prompt throughput: 3.2 tokens/s, Avg generation throughput: 1.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 0.0%
(APIServer pid=141923) INFO 02-20 13:46:15 [loggers.py:257] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 0.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 0.0%
(APIServer pid=141923) INFO:     172.18.0.2:47216 - "POST /v1/chat/completions HTTP/1.1" 200 OK
(APIServer pid=141923) INFO 02-20 13:46:45 [loggers.py:257] Engine 000: Avg prompt throughput: 3.2 tokens/s, Avg generation throughput: 1.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 25.0%
(APIServer pid=141923) INFO 02-20 13:46:55 [loggers.py:257] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 0.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 25.0%
```

</details>

---------

Signed-off-by: Chang Min <changminbark@gmail.com>
Co-authored-by: Ignasi Barrera <ignasi@tetrate.io>