Improve README: Clearer Setup, Usage Examples, and Feature Documentation for langchain-gradient

**Description**
The current README for langchain-gradient provides a good starting point but can be improved to make onboarding, usability, and feature discovery clearer—especially for first-time users of DigitalOcean Gradient and LangChain integrations.
This issue proposes several documentation enhancements to improve clarity, completeness, and developer experience.

**Proposed README Improvements**

1. *Clearer Project Overview*
Add a short introductory paragraph explaining:
- What DigitalOcean Gradient is
- What problems langchain-gradient solves
- When users should prefer this integration over other LangChain LLM providers
2. *Environment Setup Section*
Add a dedicated Environment Setup section with:
- Step-by-step instructions to generate a DigitalOcean Inference Key
- Example .env file
- Explicit note that .env should not be committed
- Example:
`DIGITALOCEAN_INFERENCE_KEY=your_access_key_here
`
3. *Installation Verification*
Add a short snippet to verify installation:
```
from langchain_gradient import ChatGradient
print("langchain-gradient installed successfully")

```
4. *Usage Examples Enhancement*
Improve clarity by separating examples into distinct sections:
Basic Invocation
- Explain `.invoke()` vs `.stream()`
- Mention expected return types

Streaming Responses
- Add a note explaining token-by-token streaming behavior

5. *Configuration Options*
Document supported or planned parameters:
- `model` (list supported models if possible)
- `api_key`
- `temperature`, `max_tokens`, etc. (if applicable or planned)

6. *Error Handling & Troubleshooting*
Add a Troubleshooting section covering:
- Missing `DIGITALOCEAN_INFERENCE_KEY`
- Invalid or expired access keys
- Model name errors

7. *Roadmap / Upcoming Features*
Replace “More features coming soon” with a clearer roadmap, for example:
- Async support
- Tool calling
- Embeddings support
- Retry & timeout configuration

8. *Contributing & Support*
Add links or sections for:
- Contribution guidelines
- Issue reporting
- DigitalOcean Gradient documentation

**Benefits**
- Easier onboarding for new users
- Reduced setup errors
- Better alignment with LangChain documentation standards
- Improved discoverability of features and roadmap

**Additional Context**
This package fills an important gap by enabling seamless LangChain integration with DigitalOcean’s serverless inference platform. Improving the `README` will significantly enhance developer adoption and usability.