feat: add a configurable category-based logger (#1352)

A self-respecting server needs good observability which starts with
configurable logging. Llama Stack had little until now. This PR adds a
`logcat` facility towards that. Callsites look like:

```python
logcat.debug("inference", f"params to ollama: {params}")
```

- the first parameter is a category. there is a static list of
categories in `llama_stack/logcat.py`
- each category can be associated with a log-level which can be
configured via the `LLAMA_STACK_LOGGING` env var.
- a value `LLAMA_STACK_LOGGING=inference=debug;server=info"` does the
obvious thing. there is a special key called `all` which is an alias for
all categories

## Test Plan

Ran with `LLAMA_STACK_LOGGING="all=debug" llama stack run fireworks` and
saw the following:


![image](https://github.com/user-attachments/assets/d24b95ab-3941-426c-9ea0-a4c62542e6f0)

Hit it with a client-sdk test case and saw this:


![image](https://github.com/user-attachments/assets/3fee8c6c-986e-4125-a09c-f5dc019682e2)

llama_stack/providers/remote/inference/fireworks/fireworks.py

@@ -8,6 +8,7 @@ from typing import AsyncGenerator, List, Optional, Union
 
 from fireworks.client import Fireworks
 
+from llama_stack import logcat
 from llama_stack.apis.common.content_types import (
     InterleavedContent,
     InterleavedContentItem,
@@ -226,12 +227,14 @@ class FireworksInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProv
             if input_dict["prompt"].startswith("<|begin_of_text|>"):
                 input_dict["prompt"] = input_dict["prompt"][len("<|begin_of_text|>") :]
 
-        return {
+        params = {
             "model": request.model,
             **input_dict,
             "stream": request.stream,
             **self._build_options(request.sampling_params, request.response_format, request.logprobs),
         }
+        logcat.debug("inference", f"params to fireworks: {params}")
+        return params
 
     async def embeddings(
         self,

llama_stack/providers/remote/inference/ollama/ollama.py

@@ -10,6 +10,7 @@ from typing import AsyncGenerator, List, Optional, Union
 import httpx
 from ollama import AsyncClient
 
+from llama_stack import logcat
 from llama_stack.apis.common.content_types import (
     ImageContentItem,
     InterleavedContent,
@@ -203,12 +204,14 @@ class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
             else:
                 raise ValueError(f"Unknown response format type: {fmt.type}")
 
-        return {
+        params = {
             "model": request.model,
             **input_dict,
             "options": sampling_options,
             "stream": request.stream,
         }
+        logcat.debug("inference", f"params to ollama: {params}")
+        return params
 
     async def _nonstream_chat_completion(self, request: ChatCompletionRequest) -> ChatCompletionResponse:
         params = await self._get_params(request)

llama_stack/providers/remote/inference/together/together.py

@@ -8,6 +8,7 @@ from typing import AsyncGenerator, List, Optional, Union
 
 from together import Together
 
+from llama_stack import logcat
 from llama_stack.apis.common.content_types import (
     InterleavedContent,
     InterleavedContentItem,
@@ -213,12 +214,14 @@ class TogetherInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProvi
             assert not media_present, "Together does not support media for Completion requests"
             input_dict["prompt"] = await completion_request_to_prompt(request)
 
-        return {
+        params = {
             "model": request.model,
             **input_dict,
             "stream": request.stream,
             **self._build_options(request.sampling_params, request.logprobs, request.response_format),
         }
+        logcat.debug("inference", f"params to together: {params}")
+        return params
 
     async def embeddings(
         self,