fix: pass csrf token on doi/manuscript fetch and clean up withCredentials

[jaredgalanis]

No description provided.

[jaredgalanis]

@markpatton withCredentials doesn't appear to be part of the fetch api. credentials is, but we are explicitly passing the csrf token here.

[jaredgalanis]

This one was missing the csrf token and was being spiked by the DOI service.

[markpatton]

Looks good. Ok. So the consequence of this is that the manuscript file fetch will just not return files until this fix is present. I think I had forgotten that functionality existed.

I tried it out and did get a 200 back from the service. But strangely the ui did not say a pdf was available despite a url I tried in the returned json resolving to a pdf. This was the json:

{
  "manuscripts": [
    {
      "url": "https://bmchealthservres.biomedcentral.com/track/pdf/10.1186/s12913-020-05409-w",
      "repositoryLabel": null,
      "type": "application/pdf",
      "source": "Unpaywall",
      "name": "s12913-020-05409-w",
      "isBest": true
    },
    {
      "url": null,
      "repositoryLabel": null,
      "type": "application/pdf",
      "source": "Unpaywall",
      "name": null,
      "isBest": false
    },
    {
      "url": "https://www.researchsquare.com/article/rs-19757/v1.pdf",
      "repositoryLabel": null,
      "type": "application/pdf",
      "source": "Unpaywall",
      "name": "v1.pdf",
      "isBest": false
    },
    {
      "url": "https://www.researchsquare.com/article/rs-19757/v2.pdf?c=1631848933000",
      "repositoryLabel": null,
      "type": "application/pdf",
      "source": "Unpaywall",
      "name": "v2.pdf?c=1631848933000",
      "isBest": false
    }
  ]
}