ODC does not shut down connection pools/threads cleanly after some types of failures

From @OrangeDog originally.

**Precondition**
- [x] I checked the issues list for existing open or closed reports of the same problem.

**Describe the bug**
Often triggered by repeated NVD failures, but appears to have revealed a number of bugs in the code, including:
```
org.owasp.dependencycheck.data.nvdcve.DatabaseException: Error updating 'CVE-1999-1186'; General error: "org.h2.mvstore.MVStoreException: Reading from file sun.nio.ch.FileChannelImpl@47993bef failed at 249080802 (length -1), read 0, remaining 512 [2.4.240/1]"; SQL statement:
SELECT id, ecosystem FROM cpeEntry WHERE part=? AND vendor=? AND product=? AND version=? AND update_version=? AND edition=? AND lang=? AND sw_edition=? AND target_sw=? AND target_hw=? AND other=? [50000-240]
at org.owasp.dependencycheck.data.nvdcve.CveDB.updateVulnerability (CveDB.java:1104)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.updateCveDb (NvdApiProcessor.java:119)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.call (NvdApiProcessor.java:96)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.call (NvdApiProcessor.java:40)
at java.util.concurrent.FutureTask.run (FutureTask.java:328)
at java.util.concurrent.ThreadPoolExecutor.runWorker (ThreadPoolExecutor.java:1090)
at java.util.concurrent.ThreadPoolExecutor$Worker.run (ThreadPoolExecutor.java:614)
at java.lang.Thread.run (Thread.java:1474)
```
```
java.lang.NullPointerException: Cannot invoke "org.apache.commons.dbcp2.BasicDataSource.getConnection()" because "this.connectionPool" is null
at org.owasp.dependencycheck.data.nvdcve.DatabaseManager.getConnection (DatabaseManager.java:576)
at org.owasp.dependencycheck.data.nvdcve.CveDB.updateOrInsertVulnerability (CveDB.java:1168)
at org.owasp.dependencycheck.data.nvdcve.CveDB.updateVulnerability (CveDB.java:1093)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.updateCveDb (NvdApiProcessor.java:119)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.call (NvdApiProcessor.java:96)
at org.owasp.dependencycheck.data.update.nvd.api.NvdApiProcessor.call (NvdApiProcessor.java:40)
at java.util.concurrent.FutureTask.run (FutureTask.java:328)
at java.util.concurrent.ThreadPoolExecutor.runWorker (ThreadPoolExecutor.java:1090)
at java.util.concurrent.ThreadPoolExecutor$Worker.run (ThreadPoolExecutor.java:614)
at java.lang.Thread.run (Thread.java:1474)
```

**Version of dependency-check used**
The problem occurs using version 12.2.2 of the maven plugin.

**Log file**
https://gist.github.com/OrangeDog/ab9ce4715fb37d5fa9d38b09dee02d83

**Expected behavior**
Graceful handling of HTTP errors and no database issues.

From @chadlwilson
This is just because the shutdown of ODC and all its threads and connection pools is not sequenced or clean.

The database is closed while there are still active threads and connections. What should happen is it should make a "fatal" decision, then interrurpt all pools/hreads/connections/workers, wait for them to complete and then close the database and exit, but it does not.

There is probably some duplicate issue here somewhere as I had some WIP to fix it. Can't find it right now. Related to #6535 though.

Start with DatabaseManager.getConnection, CveDB.updateVulnerability, and NvdApiProcessor.call from the reported stack traces, then inspect how shutdown is sequenced after repeated NVD failures. Compare the related #6535 discussion and any existing shutdown or worker cleanup paths. Done means failures are handled gracefully without active threads or connections using a closed database.

Written by the indexing model from the issue text.